US008793580B2 (12) United States Patent (10) Patent N0.: (45) Date of Patent: Robinson (54) 5,204,969 A 5,227,892 A 5,467,288 A SYSTEM AND METHOD FOR DISPLAYING AND EDITING DIGITALLY SAMPLED AUDIO DATA US 8,793,580 B2 Jul. 29, 2014 4/1993 Capps et al. 7/1993 Lince 11/1995 Fasciano et al. (Continued) (75) Inventor: Robert S. Robinson, Trenton, NJ (US) (73) Assignee: Channel D Corporation, Trenton, NJ (Us) (*) Notice: Subject to any disclaimer, the term of this patent is extended or adjusted under 35 U.S.C. 154(b) by 1592 days. Filed: EP WO 1258879 9103053 11/2002 3/1991 OTHER PUBLICATIONS Howarth, Jamie, et al “Correction of Wow and Flutter Effects in Analog Tape Transfers,” Audio Engineering Society Convention Paper, 117th Convention, pp. 1-6 (Oct. 2004). (21) App1.No.: 11/759,068 (22) FOREIGN PATENT DOCUMENTS (Continued) Jun. 6, 2007 Primary Examiner * Boris Pesin (65) Prior Publication Data US 2008/0074486 A1 Assistant Examiner * Matthew E11 (74) Attorney, Agent, or Firm * Lowenstein Sandler LLP Mar. 27, 2008 Related US. Application Data (60) Provisional application No. 60/811,249, ?led on Jun. 6, 2006. (51) Int. Cl. data recordings are digitally sampled, segmented, and ren (58) (2006.01) (2006.01) (2006.01) dered into a plurality of arc segments. A value of one or more audio data parameters are determined for each arc segment. A US. Cl. CPC . GIIB 33/10 (2013.01); GIIB 3/00 (2013.01) USPC ........................................................ .. 715/716 Field of Classi?cation Search USPC and transforming the digital samples into a user-controllable visual representation for computer-based data interpretation and editing. A plurality of synchronized streams of sampled G06F 3/00 GIIB 3/00 GIIB 33/10 (52) (57) ABSTRACT A method and system for receiving audio data, converting the audio data into discretely sampled data, such as digital audio, ........................................................ .. 715/716 See application ?le for complete search history. (56) References Cited U.S. PATENT DOCUMENTS 2,033,479 A * 5,047,999 A * 3/1936 Murphy .......................... .. 369/6 9/1991 van derMeulen ....... .. 369/30.17 visual indication of the parameter value, such as a color, hue, or shade, is determined for each arc segment. The plurality of arc segments are arranged into a plurality of arcs, and the plurality of arcs are arranged as a smooth, continuous radial spiral, or arcs, using graduated grays or colors to denote waveform characteristics to form a realistic visual depiction of a mechanical recording emulating a conventional “vinyl” record. A user may interact with the record image, using a locator cursor, to control and edit the audio data. In addition, recording defects or other features may be highlighted on the surface of the image as color keyed or iconic overlays, to assist in the editing process. 22 Claims, 9 Drawing Sheets US 8,793,580 B2 Page 2 (56) References Cited 2004/0120688 A1 US PATENT DOCUMENTS 2004/0189646 A1 2004/0199277 A1 9/2004 Hayashi et al. 10/2004 Bianchi et a1. 2004/0258392 A1* 12/2004 5,475,835 A 12/1995 Hickey 5,601,436 A 5,671,377 A * 2/1997 Sudman et al‘ 9/1997 Bleldt et a1. ................ .. 715/723 6/2004 Poltorak Morita et al. ................. .. 386/52 2005/0010409 A1 1/2005 Hull et a1. 2005/0036628 A1* 2005/0212802 A1 2/2005 9/2005 Devito .......................... .. 381/61 Takeda etal 5,675,778 A 5,682,326 A 10/1997 Jones 10/1997 Klingler 6131. 2006/0065102 A1 2006/0198610 A1 3/2006 Xu_ 9/2006 Aklfusa_ 6,100,881 A 6,320,598 B2 6,335,730 B1 8/2000 Gibbons et 31‘ 11/2001 Davis et 31‘ 1/2002 Gould 2006/0202994 A1 2007/0079228 A1 9/2006 Chevalller 6131. 4/2007 Ando et al. 6,353,510 B2 3/2002 Drouin OTHER PUBLICATIONS * giléftg?e; 6,441,830 B1 8/2002 Duvall et a1. 6,466,211 B1 6,507,349 B1 10/2002 Havre et a1. 1/2003 Balassanian 6,532,335 6,563,523 6,665,751 6’845’073 """"""""" " 715/834 B2 3/2003 Otorno et al. B1 5/2003 SUChOCkl et 31~ B1 * 12/2003 Chen et 31' B2 100% Yaimada et al' """"" " 369/5339 6’897’868 B2 * 5/2005 Chff """"""""""""""" " 345/440 Parker, Conrad, “Sweep: Audio Editing, Scrubbing and Latency Visualisation,” PP~ 1-6~ Plangent Process Software, “Frequently Asked Questions,” pp. 1-4. Van Aeken, Francis, “Jackson DJ Software Powered by Musical Metadata,” pp, 1-5, Francis Van Aeken, “Jackson: DJ Software Powered by Musical Metadata,”Van Aeken Software bvba (2005). Conrad Parker, “Sweep: Audio Editing, Scrubbing and Latency Visualisation” Linux 2003 Conference and Tutorials (Jul 31-Aug 6,958,754 B2 10/2005 Alexander et al. 3 2003 7,133,531 7,640,069 B2 B1 * 11/2006 12/2009 Karpenstein Johnston ....................... .. 700/94 WWWPlangemProceSSeSFOWfaqhtm ’ 2001/0047384 A1 11/2001 Croy 2003/0009494 A1* 1/2003 2004/0018007 A1 1/2004 Akita ’ ’ ' _ _ Weber,Marc et al. “V1sua1121ng T1me-Ser1es 0n Sp1rals”. Milne et al. .............. .. 707/500.1 * cited by examiner ' US. Patent ‘i.) Jul. 29, 2014 Sheet 1 0f 9 US 8,793,580 B2 US. Patent Fig. 2A Jul. 29, 2014 Sheet 2 0f9 . US 8,793,580 B2 32 31 30 /_/ 14 p.11 I/ I US. Patent Fig. 3A Jul. 29, 2014 Sheet 3 0f9 US 8,793,580 B2 US. Patent \ Jul. 29, 2014 Sheet 4 0f 9 US 8,793,580 B2 US. Patent Jul. 29, 2014 Sheet 5 0f 9 US 8,793,580 B2 gm 1. m US. Patent Jul. 29, 2014 Sheet 6 0f9 US 8,793,580 B2 ' Singie Eris-rape Reach-?n“ § Sm-gie Sammie Fiesgi: '~ 50610? Ems 151 US. Patent Jul. 29, 2014 Sheet 7 0f9 US 8,793,580 B2 .. x, \ \\\\\\\§§ \ \\\\ U S. Patent a i Jul. 29, 2014 Sheet 8 0f 9 US 8,793,580 B2 US. Patent Jul. 29, 2014 Sheet 9 0f9 US 8,793,580 B2 \\ ova “x mwmx \ “m \ k w: w: m é “w \ mq ‘ \\ \x §§§ \\\ $5 \ \\\\\\ \ \\ \ \ / % \ \ \\ \ \\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ \ \ \ US 8,793,580 B2 1 2 SYSTEM AND METHOD FOR DISPLAYING AND EDITING DIGITALLY SAMPLED AUDIO DATA points or boundaries of the track) based only on the appear ance of the waveform. For instance, a gradual song fade-out or fade in can be heard quite noticeably even in the presence of vinyl background noise, which may obscure the music, CROSS REFERENCE TO RELATED APPLICATION when viewed as the waveform. Accordingly, there is a need in the art for a method and system for generating an intuitive and user-friendly visual representation of discretely sampled data, wherein a user may This application claims the bene?t of US. Provisional Application Ser. No. 60/811,249 ?led on Jun. 6, 2006. The entire disclosure of US. Provisional Application Ser. No. interact with the visual representation in the form of a con ventional ‘vinyl’ record and record playback apparatus (i.e., a record player) to perform a number of tasks, including play back, editing, content management, and error/defect detec 60/811,249 is incorporated by reference herein. FIELD OF THE INVENTION The present invention relates generally to a method and system for transforming sampled data into a visual represen tation with which a user may interact. In particular, this inven tion relates to the transformation of audio data into a realistic visual depiction of a mechanical recording (e.g., a conven tional vinyl record). The present invention relates to a method tion. SUMMARY OF THE INVENTION The current invention provides an alternative means of display of information about the audio. Speci?cally, the 20 invention describes the generation of an image of an analog format vinyl record disc, used as an interactive, virtual object. of emulating the traditional playback experience of the pre digital-audio era by simulating the tactile interaction with This avoids many limitations of the current art, as well as vinyl records which were originally used as a recording and more closely and favorably linking the technical and enter tainment (such as the rotation of the image on the computer playback medium. The emulation of the visual properties of the vinyl record format facilitates the display and editing of the content of, for example, audio recordings. 25 characteristics of the display. A side bene?t to the platter image, when playing back music in a way that emulates the “album” format, is that an estimate of the remaining duration of the current track, and subsequent tracks can be made visu BACKGROUND OF THE INVENTION In the playback of digitally recorded audio, if done in 30 conjunction with a visual display, such as a computer moni 35 ear amplitude versus time waveform display. The reasons for providing the display can vary between the need for showing technical information regarding the audio and to provide an entertaining visual display (by viewing the audio waveform or frequency spectrum information, for example). 40 plurality of arcs, is used that permits ?nding features of inter est in the recording with greater precision than conventional methods, while providing an easily manipulated overview of the entire audio recording. According to an embodiment of the present invention, the plurality of digital samples are segmented into groups, or arc segments. The digital samples of each arc segment are ana lyzed to determine a value of at least one audio parameter for the arc segment. Next, each arc segment is displayed with a visual identi?er which represents the value of the at least on On the technical side, provision is usually made for manu ally altering the location of the playback position, such as using a cursor indicator on the display, controllable via input from a mouse. This is usually required for editing of the audio data, such as dividing a long recording into individual tracks. The editing is facilitated by observing visual cues in the ally. This enhances anticipation and enjoyment of the music. Instead of representing the audio as a traditional type of rectilinear waveform display, a spiral radial paradigm, or a tor, it is customary to provide some type of display that shows information regarding the audio amplitude and time offset (relative to the beginning or end of the recording) at the playback position. Typically, this takes the form of a rectilin display during playback, or applying other visual effects) 45 audio parameter (e.g., modulation). The visual identi?er, as used herein, may include, but is not limited to, a color, hue, shade, other visual characteristic which may be used to rep display, such as regions of low signal amplitude, and using resent the parameter value. This provides a user with a visual these regions as tentative locations for establishing track divi sions. One drawback to this approach is that in the display of the overall waveform of a recording, the track separation locations cannot be resolved visually, because they are typi representation in the change of the parameter in the different arc segments. Advantageously, changes in the value of the 50 parameter in one arc segment as compared to another, as illustrated by the different visual identi?ers, may be used to cally obscured by nearby audio having higher amplitudes. communicate to the user relevant information about the audio This is usually addressed by “zooming in” on a smaller por tion of the waveform, permitting the visualization of the lower amplitude audio at track boundaries. However, since the zoomed waveform only comprises a subset of the entire audio recording, a tedious scrolling operation may be required to reliably ?nd all track boundaries. An additional drawback arises when editing audio not sourced from a quiet digital recording, such as when tran content. By comparing the visual identi?ers of the arc seg ments, the user can ‘see’ changes in the audio parameter. 55 The simulation goes beyond a cosmetic, stylized rendition of the appearance of a vinyl record, because the appearance of the groove modulations re?ects the actual audio content of the recording, or possibly other parameters derived from the audio information, which also can be displayed as an overlay 60 or color shading of the vinyl image. Also, displayed in the circular format, periodic features in the recording are empha scribing an actual analog vinyl record. Here, the amplitude at track boundaries doesn’t drop to zero (digital silence); instead, a residual background noise (such as turntable low frequency noise, commonly known as “rumble”) is imposed on the quiet parts of the audio. Digital silence doesn’t exist in analog transcriptions of vinyl records, so it’s impossible to establish accurate track mark points (i.e., the start and end 65 sized, and defects such as scratches (in the case of recordings transcribed from vinyl records) used to facilitate the calibra tion of the true playback speed. According to an embodiment of the present invention, the system and method convert discretely sampled data into a display that emulates the vinyl record format. Then, the famil US 8,793,580 B2 4 3 iar tonearm/stylus/vinyl record metaphor can be used for the ?rst time as a tool for editing and playing back digital audio ?les. For example, inter-track silences are rendered as plainly visible areas of low modulation, appearing as discrete circular FIG. 8 illustrates a comparison of the performance of exemplary approaches for locating physical defects during a calibration process, according to an embodiment of the present, invention; and FIG. 9 illustrates an exemplary process for generating a bands, rather than being compressed visually and obscured by platter image, according to an embodiment of the present adjacent high amplitude areas of the audio signal. This pro invention. It is to be understood that the attached drawings are for purposes of illustrating the concepts of the invention and may vides a visually informative cue or track mark starting loca tion (i.e., a starting boundary of the track). The vinyl record image waveform display format further expands this meta not be to scale. phor, because by manipulation of the computer input device, DETAILED DESCRIPTION OF THE INVENTION such as a mouse, the playback position can be manually ?ne tuned by “grabbing” and “spinning” the vinyl disk, while The present invention relates to a method and system for simultaneously listening to a loopedplayback; of a fraction of a second’s worth of audio. generating a visual representation of input audio data After navigating to a speci?c place of interest in the audio recording with the aid of the vinyl image, which is a primary received from a source, wherein in the visual representation emulates a conventional vinyl record. The input audio data advantage compared to an overview type rectilinear wave form display, the process also may be enhanced at this stage by viewing a highly magni?ed or zoomed version of the waveform, as a visual overlay, in the familiar rectilinear for mat. In this way, the two methods of displaying the recording are complementary and reinforce each other’s utility, while may be in either analog or digital format. If the input data is in analog format, the analog data is ?rst converted into a plural avoiding the tedious task of having to scroll slowly through the recording using only a zoomed in rectilinear display. Setting track marks (i.e., the boundaries of the track) inter actively using both the waveform and audible feedback elimi nates the possibility of inadvertently placing a track mark 20 digital format and comprise a plurality of digital samples, and, thus, no conversion is required. The plurality of digital samples (either as received from the 25 before the actual fade-out or after an actual fade-in. The present invention allows a user to intuitive grab and spin the 30 source or as converted) are then segmented into a plurality of arc segments. Next, for each arc segment, the value of at least one audio parameter is determined. The arc segment is then rendered and displayed with a visual identi?er which visually represents the value of the at least on audio parameter. The visual identi?er may be a color, shade, hue or other visual expression of the value. By presenting each arc segment with a visual identi?er representation of the value of the selected audio parameter, the changes in the parameter may be seen when viewing the plurality of arc segments when arranged “platter” to re?ne and accelerate the editing process. The general familiarity of the public with such records and their associated playback equipment is an advantage, as most persons already possess an intuitive grasp of the concept of the vinyl LP disc. For users lacking familiarity with analog ity of digital samples, according to any suitable method known in the art. Alternatively, the input audio data may be in 35 into a series of arcs, the series of arcs emulating a record turntables and vinyl records, these elements present an attrac tive aspect of the design, given the current resurgence of image. interest in this recording and playback medium. resentation of the input audio, herein referred to as the “record A plurality of the arcs are combined to form a visual rep image”. The record image comprises a plurality of arcs, BRIEF DESCRIPTION OF THE DRAWINGS 40 The present invention will be more readily understood from the detailed description of exemplary embodiments pre sented below considered in conjunction with the attached drawings, of which: FIG. 1 is a diagram of an exemplary data display including characteristics of a conventional record playback apparatus, according to an embodiment of the present invention; FIG. 2A illustrates exemplary components of the data dis play, according to an acoustic-model data rendering embodi ment of the present invention; FIG. 2B illustrates an exemplary components of a data exemplary record image generated according to the present 45 invention. Advantageously, a user may interact with the record image much in the way one interacts with a conven tional vinyl record to perform a number of functions, as described in detail below. The digital samples of the input data are processed (as 50 described below with reference to FIGS. 2A and 2B) and converted to the radial representation, or plurality of arc seg ments making up the larger arcs of the record image. This may be accompanied by visual feedback of the ongoing process, display, according to a physical-model rendering embodi ment of the present invention; FIGS. 3A and 3B show modi?ed data renderings using subsets of the data shown in FIGS. 2A and 2B; arranged to emulate a conventional “vinyl recor .” Embodiments of the present invention are described below in detail with reference to FIGS. 1-9. FIG. 1 illustrates an denoted by progress animation arrow 4, as the image data, is progressively calculated and overlaid on the platter substrate 55 3. Portions of the digital samples with large amounts of modulation, as assessed by the analysis algorithm, are dis FIGS. 4A and 4B illustrate a process according to an played as image highlights 1, while low levels of signal embodiment of the present invention wherein a radial depic tion of waveform data is used to assist in locating features of invention to locate track boundaries in an analog music modulation 2 are represented as unchanged, or nearly so, compared to the substrate data display area 3. The platter substrate 3 may be displayed as dark gray or black color, or as a solid, bright color. The platter substrate 3 recording; may also be patterned for aesthetic, ornamental purposes, interest in the sampled data ?le; 60 FIG. 5 illustrates an use of an embodiment of the present FIGS. 6A, 6B, 7A and 7B illustrate an exemplary process for calibration of the time base of a data sample using physi cal, periodic defects present in the source material, according to an embodiment, of the present invention; 65 such as with a design, photographic image, or other illustra tion. The highlights may be drawn with a variable opacity from 0 to 100 percent, with a 100 percent value obscuring the image of the substrate. Low levels of opacity may be used for US 8,793,580 B2 5 6 aesthetic enhancement of the display, in conjunction with for aesthetic compliance and conformity with the physical playback medium being emulated. different substrate colors or visual patterns. The highlighting in areas with high waveform modulation, and substrate prominence in areas of low modulation may be inverted, providing a negative shaded image. The modulation may be As shown schematically in FIG. 9, beginning at the lead-in area 7, the image data information is applied to the blank platter substrate 3. Each pixel in the image is treated as a sub-segment of a larger arc, and has a variable, diminishing (in the case of image data application begun at the lead-in represented as gradations of gray tones or as false-color shad ing. A combination of the two may be used to convey addi tional information in the data display. For example, color shading might be used to indicate differences in relative amplitude or phase between a plurality of channels. Other aspects of the data display, which is con?gured to area) radius. The sub-segment is herein referred to as an arc segment. As such, according to an embodiment, of the present invention, each pixel equates to one arc segment. As the image data is applied, the arc radius is diminished. The effec emulate a familiar object, an audio recording playback tum tive radius is calculated for each pixel of the image. The radius table, include a label area 7 for various information, a radial need not have a whole-number value, because modern com spindle 6, tone arm 5, playback cartridge 9, playback stylus 8, puter graphic imaging programs and routines are con?gured to alias intermediate, ?oating-point representations, thus pro cueing emulation button 10 and lead-in area 11. According to a preferred embodiment of the present invention, a linear style carriage-type tone arm is shown; but other aesthetic variations may include pivoted straight or curved tone arms. A linear design is illustrated in the ?gures because of the viding increased realism of the spiral image drawing. operations, as described in detail below. For example, as shown in FIG. 9, given a substrate 1 with radius 2 of 820 pixels, and a lead-in radius 3 of 800 pixels, the ?rst pixel applied is considered to be part of an arc segment 4 having a radius of 800 pixels. The shading (brightness or One having ordinary skill in the art will appreciate that features 5, 6, 7, 8, 9, 10, 11 are optional, and may or may not be included in the data display. These features, used here as a analysis model, as explained below. In the case of emulating an analog playback disc, the next pixel, arc segment 5, is simpler computation of data offsets during emulated cueing functional aesthetic construction, are intended to emulate components, features and aspects of a traditional audio ana 20 color) of this pixel (or arc segment) is determined by the 25 log disc recording playback system (turntable). Embodi ments of the present invention incorporate these elements to leverage the user’s likely familiarity and comfort level with this particular object (i.e., the turntable). For users lacking familiarity with analog turntables, these elements present an attractive aspect of the design, given the current resurgence of 30 35 The brightness or color of the image is calculated at a 40 The image construed on may commence at any location on the substrate, or even at the innermost radius of the substrate. 45 location is chosen a small distance inset from the outer simu to maintain a spiral appearance. According to an embodiment of the present invention, each radius step employs a ?xed-radius, circular arc; each revolu non-interconnected circles. This design allows the inclusion of many of the desirable characteristics of the record image, according to an embodiment of the present invention. A pre able-radius, noncircular, spiral arcs to construct the record image. 50 The label area 7 may have a radius between 5 percent and 90 percent of the substrate radius, although the optimum value would be in conformance with the physical medium emulated, such as, for example, a 7 inch diameter 33 or 45 55 The arc length also affects the way the input data is ana lyzed. The input data is segmented into an integer number of digital samples per arc segment. The optimum arc length for emulation of an analog playback disc is determined by the disc emulation model rotational rate, in conjunction with the sample rate of the digital input data. This arc length is deter mined by the following relation: where s is the angular rotational rate of the disc, and 60 Fs the digital signal sample rate. For example, given a sample rate of 44.1 kHZ and a disc rotational rate of 331/3 revolutions per minute, each digital sample occupies an arc angle of (360 degrees/revolution)* somewhat diminished image data display capacity, and may be useful in certain other contexts. A small band of the substrate adjacent to the outside radius of the label area may be reserved for the lead-out area, again the ?nal image, balanced against the computational time ferred embodiment, of the present invention employs vari lated edge of the substrate, commonly known as the lead-in RPM, physical recording disc; a 10 inch 33, 45 or 78 RPM physical recording disc; or a 12 inch 33 or 45 RPM physical recording disc. For purposes of illustration of an embodiment of the present invention, a 12-inch 33 RPM LP format with multiple individual music tracks is shown, with a label radius approximately 20 percent of the substrate radius. This is somewhat less than normally used with a physical analog disc. The present invention also lends itself to construction of single-track 12, 10 or 7 inch physical format emulation, for a According to an embodiment of the present invention, the unit of length of the arc segment is expressed in degrees. The arc length (in degrees) is determined by the desired quality of tion of the generated image consists of concentric, discrete, However, in accordance with the aesthetics of the emulation area 11. A portion of the image display area near the inner radius also is reserved for a legend, printed description or decorative image or design, the label area 7. 5 depends on the circumference and radius of the spiral arc decreased by ((2 pi)/360) pixels for each segment to continue at a point lying somewhere on the substrate. of the familiar analog disc playback paradigm, a starting tion). The starting radius 6 of the next arc segment (or pixel) required. For example, if a ?xed arc length of 1 degree is selected, the radius of the arc also must be continuously plurality of points. The practical limit of the number of points or pixels in the image is determined by the speed of the host computer and the resolution of the display device. Regardless of the resolution chosen, the image construction commences analog disc normally is spun in a clockwise fashion, so increasing time coordinate is in the counterclockwise direc tion; the image data could also be applied in a clockwise direction in an alternative embodiment of the present inven being considered at that point. interest in this recording and playback medium, even among the demographic born after the onset of the mainstream appli cation of digital sound recording. applied counterclockwise from the ?rst, pixel (because an 65 ((33+1/3 revolutions)/ 60 seconds)/(44100.0 second):0.004535 degrees per digital sample. samples/ Given the above parameters, the arc segment length must therefore be constrained to multiples of 0.004535 degrees. At US 8,793,580 B2 7 8 an arc spiral radius of 800 pixels, this corresponds to an arc the art will appreciate that, in practice, the tradeoff between drawing many small arc segments and computational ef? ciency dictates that arc segment lengths of greater than one segment circumference (length) of 0.06332 pixels per data sample. One having ordinary skill in the art will appreciate that the input data could be progressively resampled to any practically attainable sample rate, generating the optimum number of sampled points for a given arc segment length. pixel (including more data samples per arc segment) and arc Sine widths greater than one pixel be used. As such, according to an embodiment of the present invention, a typical arc line width of square root (2) pixels is used. For large data sets the number of samples per arc segment, According to an embodiment of the present invention, a minimum arc length of 1 pixel is considered. In the above example, a minimum arc segment length of 1 pixel corre spond to 1/0.06332 or 15.79 data samples. Since an integer can be increased and/or the arc line width decreased. These parameters are adjustable at the discretion of the user, to provide the most aesthetically pleasing image, while main number of samples is required, this ?gure is back-calculated using a minimum value of 1.6 samples per analysis sample, taining a reasonable computational rate. For example, gener giving a segment length of 16/15.79 or 1.013 pixels. minutes of sampled digital audio on a currently shipping consumer-level computer workstation takes approximately ating a complete, high quality spiral image “platter” from 30 Therefore, the arc segment length is predetermined by the sample rate of the input data. As the spiral radius decreases, the arc segment length, in pixels, also decreases, in proportion to the radius. Therefore, to maintain the minimum design constraint of 1 pixel of arc length, the number of samples per segment must be gradually increased (because the arc angle 30 seconds. According to embodiments of the present invention, two primary signal analysis models may be used to emulate the appearance of the record image. One having ordinary skill in 20 image where areas of differing signal characteristics can be differentiated upon visual inspection of the image. The visual A computational shortcut may be taken at this juncture. Arc segments with lengths greater than one pixel may be applied that have a ?xed radius within the segment. These ?xed radius segments are then joined to a previous segment having a representation may be based on one or more of the following 25 slightly larger and a following segment having slightly smaller radii, respectively. The granularity caused by this 30 per minute value; results of signal convolution showing coherence with a comparison signal; and other known signal characteristics. Although one having ordinary skill in the art will appreciate that the present invention may be con?gured 35 parameter, for the purposes of illustration, the exemplary 40 embodiments described herein related to the present inven tion are described with reference to signal characteristics/ parameters described herein as the level of amplitude of modulation. FIG. 2A shows a record image producing according to an One additional step was performed to reduce the promi nence of the locations where arcs are joined. The Root Mean Square (RMS) values (explained below) obtained are slightly low-pass ?ltered, so that the change in highlighting from one segment to the next is less abrupt. The ?ltering is a simple to generate a visual representation of any suitable signal ?rst-order In?nite Impulse Response (IIR) ?lter function, h0:h1 Equation 2 exemplary signal characteristics, including, but not limited to the interchannel or single channel phase or amplitude (modu lation level); frequency balance; signal amplitude at a par ticular or range of frequencies; total harmonic or intermodu lation distortion over a range of or at a single frequency; beats method is practically invisible. This technique was used to generate the images included in the Figures. the art will appreciate that alternative models similar to the ones described in detail herein may be used to create a record must be increased). This causes discrete changes to the arc segment lengths, that were found to be unnoticeable. where h1 is the highlighting parameter applied to the current exemplary model according to an embodiment of the present segment; invention, herein referred to as the “Acoustic” model. Accord h0 is the highlighting parameter applied to the previous seg ment; and ing to this embodiment, the Acoustic model calculates the c1 is the ?lter coef?cient. According to a preferred embodiment of the present inven 45 as described in detail above. The input signal typically com prises two channels (stereo), in the case of an audio music tion, c1 has a value between 1.0 (no ?ltering) and 0.01 (sig ni?cant ?ltering), with a value of 0.9 determined to be opti mum. After calculating h1, its value is substituted for h0 which then becomes the previous segment’s highlighting 50 55 60 calculated highlight level. For example, at high levels of modulation, the opacity may be increased to approximately 90 percent, and reduced proportionate to the modulation level to a minimum of approximately 30 percent at locations of low or zero modulation. Thus, if the substrate blank; color is a aliasing and transparency of the line segments, provided by the host computer’s built-in graphics routines, may be dark; blue, the highlights appearbluish white, and the areas of adjusted to cover gaps in between adjacent arcs at different radii. According to a preferred embodiment of the present invention, the arc segment length may correspond to the drawn width of the arc segment. One having ordinary skill in the substrate were so imprinted. According to an embodiment of the present invention, the opacity of the arc drawing may be varied depending on the samples per arc segment) and arc line widths greater than one pixel be used. According to an embodiment of the present invention, a typical arc line width of square root (2) pixels is used, and a radius step of 1.0 pixel per revolution. Line recording. However, any number of channels, including addi tional channels, may be included in the analysis. The high lighting amount (i.e., the pixel brightness) applied is propor tional to the computed RMS value for the data sample. At lower amounts of highlighting, the opacity of the arc drawing may be reduced proportionately, to allow the color of the substrate to show, or a decorative design to show through, if value for the next iteration of the arc rendering. Note that such highlight smoothing is not a requirement for the present invention, but may optionally be applied to improve the appearance of the record image. In practice, the tradeoff between drawing many small arc segments and computational ef?ciency dictates that, arc seg ment lengths of greater than one pixel (including more data RMS amplitude of the sum of the synchronized (in time) input signal channels, for the number of samples per arc segment, 65 low modulation bluish black (black being the arc color used for areas of low modulation). The preferred variable opacity used is between approximately 5 and 1.00 percent. Altema tively, the opacity of the overlaid arc drawing may be main US 8,793,580 B2 9 10 tained at a ?xed value between 5 and 100 percent. At 100 data offset is calculated by the rotational rate represented by percent opacity, the appearance of the image would depend the platter image times 1/360 times the manually changed angle of the platter. solely on the arc drawing and would not be affected by any coloration or patterning in the substrate. FIGS. 2A and 2B illustrate an exemplary embodiment of the present invention. As shown in FIGS. 2A and 2B, portions of the data with low signal modulation appear as a dark band 11 in the image. Areas with moderate or high modulation become highlighted according to the level of modulation, as An alternate method of determining an accurate offset into the source ?le may be accomplished by saving a lookup table with an offset corresponding to each rendered image point, or a lookup table for each image radius, and the data offset calculated based on the sample offset for a given offset angle from the lookup value. The precision in generating the image is suf?cient to ensure pixel-accurate correspondence between 12. Iconic markers indicated by 13 and 36 highlight regions of interest, and are superimposed on the image. Here, the mark ers are con?gured to indicate putative transients in the data, the image and the corresponding original sampled data. In the case of a more complicated “vari-pitch” image generation method mentioned below, the arc radius would not necessar caused by defects (pops) in the source (digitally sampled from an actual analog record platter). The algorithmic method for pop detection in conjunction with the data display is ily decrease in a simple linear fashion during the generation of the image, and an alternate method, such as a look-up table, may be used to correlate the stylus position and data offset. described in detail below. Markers also can be displayed as a The angular position of the platter is controlled by clicking circular highlight, as 42. The lead-in area as explained above is indicated by 14. In a and spinning the platter, in emulation of the familiar turntable preferred embodiment of the present invention, additional 20 paradigm. A “hand” cursor 33 is used to provide a feedback cue for the user. One having ordinary skill in the art will parameters are adjustable; a proportional slide control for appreciate that any suitable pointer icon may be used in the make-up gain 15. A Repeat parameter 16 used in conjunction present invention. The platter-spinning paradigm and its applications to examining and editing the data are explained with an Editing feature and settings con?gured with controls 18, 19, 20 is detailed below. “Stylus cueing” for the emulated turntable is provided by control 17; playback signal ampli 25 tude metering 28 and monitoring volume adjustment 29. Con trols 30, 31, and 32 affect the operational mode of the pre ferred embodiment of the present invention; namely, mation display on the label area 34, including artist name 37, title of recording 38, track names and times 41, plus spaces for playback, editing or archiving (recording) mode, respec tively. 30 “offset” refers to the position in number of digital samples with reference to FIGS. 6 and 7. The additional information 43 may include the date of the recording of the digitally 35 cal model) is indicated by 46 and 47 on the label data area, according to an embodiment of the present invention. The data. For an audio recording, this could be represented either Physical rendering model generates a somewhat different 40 image (shown in FIG. 2B), than theAcoustic model (shown in FIG. 2A). The overall difference between the images gener data offset time coordinate in minutes and seconds is indi cated by time display 27. To assist in locating a low-modula tion area, a ribbon display 26 representing the integrated highlighting at each discrete radius is provided. According to an embodiment of the present invention, the ribbon display sampled music or data ?le. The rendering model used (i.e., the Acoustic or the Physi from the beginning of the recording of digitally sampled input by the sample number or by a temporal value (time coordi nate) in seconds. The exact sample position is indicated by stylus 23; sighting aids are provided as marks 22 and 24. The additional data 35 and 43. The additional information 35 may include the calibrated platter rotational rate/pitch adjustment, the application of which is described in greater detail below In accordance with the tumtable/platter paradigm, the off set into the digitally sampled input data can be adjusted by moving the emulated cartridge 21 attached to the emulated tone arm 25. As known in the art and used herein, the term below, in conjunction with FIGS. 4 and 5. Optionally, based on the type of input data, additional features may be added to the record image. For example, for a digital music recording, the record image may include infor ated by the two models are not limited to contrast and/or brightness differences in the generated highlighting. This is 45 illustrated by the arc highlight indicated by 48 in the Acoustic model and 49 in the Physical model. The prominent highlight 48, at the same radial offset indicated by 49, illustrates an represents the mean amplitude value of the signal over one example of the kind of differences in the image appearance circular arc (one revolution) at the radius on the platter image corresponding to the radial position on the ribbon. Its purpose which result from the choice of the Acoustic or Physical model. Other differences in the models may be found in is to provide an additional visual aid to locating areas of low 50 comparing the images of FIGS. 2A and 2B. The Physical model is designed to more closely emulate the physical appearance of an analog recorded disc. The translation of an electronic signal to the physical undulations 55 when the stereo channels have a reverse polarity relationship. or high modulation, for manually adjusting the playback or editing location with the emulated stylus/cartridge. Although the ribbon is con?gured here to show the signal amplitude/ modulation level, it alternatively may be con?gured to dis play other suitable signal parameters. on the disc causes a greater physical undulation to appear The stylus radial offset from rest position at the lead-in area Therefore, to emulate the physical appearance of the disc, the (data offset time coordinate 0) and angular position of the Physical model subtracts the corresponding digital samples platter are used to back-calculate using an inverse of the of the stereo channels before calculating the RMS amplitude image generation algorithm to generate an accurate offset into the digital source data ?le used to generate the image. For value. In practice, visual comparison of actual, physical plat 60 example, given a manually chosen stylus position, the offset into the data is simply the fraction of the total radial displace Other models could be constructed, such as using Peak ment from the lead-in area to the start of the lead-out area, because each revolution of the platter represents the same amount (time coordinate) of data (at constant rotational velocity). When spinning the platter manually, such as when editing the sampled data, as described below, any additional ter recordings to the emulated images usually yields the most realistic representation when the Physical model is used. 65 waveform values to generate highlighting information, for example. However, in the preferred embodiment of the present invention, the best results in generating interesting, informative and aesthetically pleasing images were obtained with the two models described herein. US 8,793,580 B2 11 12 ner frequency between the two regions being approximately 1 kHz. The corresponding playback equalization is the inverse An additional aspect of FIGS. 2A and 2B is that the entire sampled data ?le was used to generate the platter image. Here, the sampled ?le was a continuously recorded digital tran scription of two sides recorded from a vinyl analog music of the curve used in the disc manufacturing process. The de-emphasis applied at playback to high frequencies mini mizes the in?uence of high frequency noise generated during the playback process. The low frequency emphasis compen disc, Creedence Clearwater Revival’s “Cosmo’s Factory,” Mobile Fidelity catalog number MFSL-l-037. An accurate emulation of the original physical platter would consist of sates for the low frequency roll-off applied to the sound only one side of the music disc. In the Edit mode of the recording during cutting of the disc, to limit the mechanical preferred embodiment of the present invention, the full-?le platter image assists in selecting the individual track mark locations. For example, using the track editing features of an embodiment of the invention, described below, the locations in the digitally sampled recording corresponding to Side 1 and Side 2 of the original, physical vinyl based recording are excursion of the disc cutter, which is greatest at low frequen cies. A strict recreation of the physical characteristics of the disc would apply the exact RIAA emphasis/de-emphasis curve. Both platter generation models used in the present invention use a hybrid approach that only attenuates the low frequencies below 100 Hz, with a single-pole roll-off similar to the RIAA equalization scheme. The high frequencies are established, as are the individual track or song locations/ left emphasized, which produces a satisfactory result. Changes in appearance of the platter image naturally would offsets, by visually locating areas of low modulation on the platter image, and manually positioning the stylus 23 at each of these locations, in turn, and noting the corresponding sty lus positions. In practice, the stylus position coordinates would be noted and saved by the software application hosting 20 the invention, at the command of the user. This process is further explained below in the description of FIGS. 4 and 5, and in greater detail below. After assignment has been com pleted, the individual emulated disc side platter images are then generated from the corresponding subsets of the ?le. The manufacture of analog music discs sometimes result from different ?ltering schemes. However, the choice of a particular ?ltering scheme is not fundamentally required by the present invention. In FIG. 2B, the segment indicated by the double arrow 50 represents the digital samples from side 1 of the sampled music disc; the double arrow of 51 represents digitized infor mation front side 2 FIG. 3 indicates the Play mode 58 of a employs a technique known to practitioners in the art as preferred embodiment of the present invention, after gener ating the individual disc side images. The Play mode loads in the disc side images and adjusts the sensitivity of the stylus “vari-pitch,” which adjusts the inter-groove spacing (pitch) of positioning (time coordinate) accordingly. The image seg the disc. This prevents areas of large modulation from causing the cutter head, used to generate the master stamper disc, front crossing into a previously cut groove, ruining the stamper. 25 30 the platter image data, 54, 55, 56, 57 also is updated accord The inter-groove spacing also may be controlled manually at the discretion of the mastering engineer. Normally, inter groove spacing is smaller on quiet areas of the disc and larger on loud areas of the disc, particularly those with high ampli tude low-frequency program content. This technique gener ment 50 of FIG. 2B corresponds to the image segment 52 of FIG. 3B. The image segment 51 of FIG. 2B corresponds to the image segment 53 of FIG. 3A. The label information area of ingly for side 1 in FIG. 3B and side 2 as shown in FIG. 3A. 35 FIG. 4 illustrates the use of the platter cueing paradigm to adjust the offset of the waveform inspector. FIG. 4A shows an offset into the original ?le, obtained by clicking and sliding ally increases the duration of audio that can be placed on a the cartridge and stylus 62 to the desired offset. The mouse is disc, compared to using a ?xed inter-groove spacing dictated by the maximum modulation level of the recording. positioned above the image of the platter, and provides feed According to an embodiment of the present invention, the back to the user by clenching the hand cursor when the mouse 40 is clicked. At this stage, the preferred embodiment of the present invention reveals a waveform display, indicating the source waveform represented by the platter image at the offset of the stylus position. Here, the offset has been adjusted to 45 waveform amplitude maximum; in this case, caused by a method and system employ a ?xed inter-groove spacing. Consequently, visual comparison of platter images created by the systems and methods of the present invention and corre sponding physical media (if transcribed digitally from an analog disc) illustrate the differences that exist therebetween. place the stylus over an iconic overlay 62 that indicated a However, there are a plurality of different results possible physical defect (pop) on the analog source disc. The corre when mastering the physical recorded disc, as dictated by the judgment of the mastering engineer. Because of this uncon sponding time offset in the source data is indicated by display 60. The waveform 63 is comprised of left channel 64, right trollable variable, the platter image generated by the method an system of the present invention resemble, but not neces 50 sarily appear identical, to a physically manufactured product made using the same audio data. While it would increase the complexity of the platter image generation model used by the present invention, it would be feasible to apply similar vari pitch or adjustable inter-groove spacing techniques in the 55 invention. The models used to generate the platter image use nearly untiltered digitized input data, which, when obtained from samples of analog music discs, has already been equalized to compensate for the emphasis scheme used for playback of analog disc recordings. Here, nearly un?ltered indicates that position 70 to new position 71, in the direction, illustrated by arrow 72, rotating the platter image clockwise about the cen ter spindle, and incrementing the offset into the data ?le. This is indicated by an increase of approximately 10 milliseconds in the offset time indicator 69, the change in position of waveform maxima icon 74, and translation of the peak 68 from waveform 63 by distance 73 in the waveform display. FIG. 5 illustrates using the platter paradigm to determine and set audio recording track boundaries. This may be accom 60 the input data samples are ?ltered to less than the usual extent dictated by the pre-emphasis signal ?ltering that’s normally applied during the manufacture (during the mastering stage) of vinyl records. For example, the RIAA equalization empha channel 66 and 67, and their normalized sum 65. In FIG. 4B, the mouse has been dragged, from former plished visually using only the waveform inspector 84, or visually and audibly with the inspector in conjunction with listening to a de?ned, continuously looped portion of the audio ?le of interest. 65 In FIG. 5, the stylus is positioned in the platter lead-in area, just prior to the start of the music information. The waveform sis curve, well-known to practitioners of the art, accentuates inspector display is split into two portions. The left half, 76 is high frequencies while attenuating low frequencies; the cor the waveform at a time offset prior to the stylus position. The US 8,793,580 B2 13 14 right half, 77, depicts the waveform at a time offset following the stylus position. The ?ducial mark 83 indicates the wave embodiment of the present invention, these in?uences are lumped together and considered to be due to turntable abso form at exactly the stylus position. lute speed inaccuracy. Each half of the waveform display is independently nor malized for amplitude. The waveform halves depicted in 76 and 77 are halves of a contiguous waveform; the apparent platter, surface noise caused by physical damage to the disc In the case of sound data sourced from an analog recording surface, due to normal wear and tear, tend to accrue. Some of discontinuity is caused by differences in scaling applied to the display. The waveform immediately to the right of 83 appears this noise may be caused by scratches or physical contami nation involving adjacent grooves on the analog disc. The noise is easily identi?able by its sound as an audible “pop” or as a prominent transient in the waveform display. The peri odicity of such pops in two adjacent grooves is approximately equal to the reciprocal of the disc rotational rate. For a 331/3 RPM disc, this would be 1.800 seconds. Any deviation from smaller because its scaling is in?uenced by the onset of the music waveform at 78. As the mouse is clicked and dragged on the platter image surface, the waveform in the display 84 scrolls horizontally and is rescaled in two halves about the ?ducial point 83. (The this value would re?ect an error in the turntable playback rotation rate. The measured deviation can be used to recalibrate the waveform depicted comes from the same source used to gen erate all other Figures.) The turntable platter paradigm becomes extremely useful in setting a track mark point, especially when dealing with image generation to increase the realism of the platter image simulation, and also for correcting the time base (absolute data sampled from an analog source. In contrast to data origi nating from a digital recording, analog data often is accom 20 pitch) of the digitized recording via resampling. Techniques panied by various forms of background noise. Unfortunately, for resampling digital audio to arbitrary values for pitch because of the masking effects of the noise, it’s not always modi?cation are well-known to practitioners of the art. The possible to accurately determine the beginning or end of an audio track based solely on the appearance of the waveform. mining the degree of pitch correction required. In the preferred embodiment of the present invention, the Mark-In mode selected by pressing control 81 causes the audible playback and continuous looping of the waveform from the edge of the frame 85 to the ?ducial 83, the portion of the frame denoted by 76. The duration of the loop is set by the Repeat interval control 16 in FIG. 2A, here 100 milliseconds. The track mark-in, or start point of the track, may be calibration procedure described below is valuable for deter 25 30 precisely determined by gently rotating the platter, which sets the precise stylus offset, while listening to the playback. The platter is rotated until any audible lead-in to the music wave form 78 is absent. The auto-normalization of the lead-in waveform also applies to the audible data as well as the waveform inspector. This ampli?es the quiet prior to the music introduction, ensuring that any musical information, is included within the track mark-in, even if masked by noise, and the nonmusical portion of the recording is excluded. In FIG. 6 a sorted list 89 and 99 of the amplitude maxima in the digitally sampled ?le is presented. Two adjacent entries 35 in the list at time offsets 26:10.90964 (90) and 26: 12.69802 (100) are separated by 1.78838 seconds. This is close to the putative turntable rotation period (for a 331/3 RPM 12" LP record) of 1.8000 seconds per revolution, and the maxima do indeed correspond to a “pop” or defect on the surface of the source analog disc recording. (There is an additional maxima, at 26:09.12041 seconds that is indicated on the platter image iconic overlay 92; but the calibration example below focuses on the other two maxima. The time delta between maxima 91 and 92 is 1.78923 seconds, therefore the percent relative error between choosing among these two measurements for cali bration is 100*(1.78923—1.78838)/1.78838 or less than 0.05 40 percent.) When a satisfactory mark-in has been established, it may be In FIG. 6A the stylus is positioned at the ?rst maxima at time offset 26:10.90964 seconds. The maxima also is indi ?nalized, in the preferred embodiment of the present inven tion, and displayed accordingly in list 82. platter, the offset may be ?ne-tuned to coincide with the peak A similar procedure is used to establish the end point of the track, also referred to as the track mark-out position, except that the mark-out mode 88 is selected, and the looping mode of the inspector display is reversed. Instead of looping the portion of the waveform prior to the cursor position, 76, the part of the waveform looped during playback is that after the cursor position, between ?ducial 83 and edge of the looping frame 86. In a similar fashion to that described above, the platter is rotated until musical information at the lead-out of the song is absent. This is ?nalized and used as the Mark-Out as depicted 87. FIGS. 6 and 7 depict using defects in the recorded material to calibrate the proper playback speed. A primary source of cated iconically on the platter image 91. By rotating the 45 be chosen. The sum of the waveforms of the two channels also 50 determine the rotation rate. The line frequency of utility power is subject to variation, which affect the rotational speed accuracy. Mechanical tolerances in the turntable components can also affect the rotational speed. Finally, playback speed inaccuracy can arise in the case of sampled digital audio if the sample clock rates of the recording and playback devices are different, again due to component tolerances. According to an is displayed 94. The selection of the peak maximum may be done manually or automatically. In a preferred embodiment of the present invention, the selection of the ?rst calibration offset is con?rmed by clicking button 96. The time offset is echoed in the text display 97. The next maxima 99 is selected in the list and ?ne tuning of 55 waveform maxima position 104 performed manually, if nec essary. The iconic representation 91 of the ?rst maxima at time offset 26:10.90964 (89) has rotated clockwise to 91' and the second maxima at time offset 26:12.69802 now is posi error in transcription of analog disc recordings is the quality of the speed accuracy of the turntable. Many mid-priced “audiophile” turntables rely on an AC synchronous motor to maximum 95 (right channel) or 93 (left channel). Generally, the channel with the transient having the most consistently prominent waveform shape among the two time offsets would 60 tioned (102) directly under the stylus. If the rotation rate of the turntable were exactly 331/3 RPM, the iconic overlays 92, 91/91' and 102 would be positioned on adjacent arcs of the platter image, instead of being offset circumferentially from each other. The offset occurs on the image because of the turntable rotational velocity error. Con?rmation of the second 65 calibration mark is con?rmed by pressing button 106, and the corresponding time offset 107 and calculated actual rota tional rate 108 are echoed on the display. Pressing button 101 US 8,793,580 B2 15 16 con?rms the calibration and regenerates the platter image, bly by playback). (Event 125 was generated by lifting the physical playback stylus from the disc, understandably gen erating a large amplitude transient.) The second algorithm basing the platter revolution on a period of 1.78838 instead of 1.8000 seconds. The resultant platter image show’s that the iconic overlays uses the maximum slew rate of the left or right signal chan nels. Of the 22 candidates displayed in the list 121, only one indicating the peak maxima are now adjacent, as shown in FIG. 7B (109), which focuses on the iconic overlay detail. FIGS. 7A and 7B are described more thoroughly below, how (122) was an actual pop event. The other candidates were comprised of valid musical information. Noticing that the transient waveforms in FIG. 6A showed that the relative signal polarity during the pop event was ever, overlays 112. 111' and 110" directly correspond to 92, 91/91' and 102, respectively, in FIGS. 6A and 6B. This calibration procedure could conceivably be applied at inverted at the peak of the pop, another algorithm that mea sured the difference between channels was used. In list 120, 13 of 22 candidates highlighted (130) were veri?ed as being different regions of the recording, in case the absolute rota tional error varies throughout the recording process, and pre suming that other surface defects exist at advantageous loca caused by physical defects in the analog disc. The other pop tions on the recording. However, it’s unlikely that properly cared-for analog discs will have a large number of physically event, 129 was the stylus lift mentioned above. This event also has characteristics in common with pop defects, namely the suitable defects; therefore, this technique is primarily large amplitude inverted polarity difference between chan intended as a means of a single-point rotational rate calibra nels. While this invention has been described in terms of several tion that’ s applied uniformly for the duration of the recording. It is possible that over a time period of typically 30 minutes, representing the duration of a single side of an analog disc, the preferred embodiments, there are alterations, permutations, 20 short-term variation in absolute rotational rate error can be neglected. For example, another suitable pop defect was located on this recording with the aid of automated tools. In FIG. 8 at offsets 33:09.88402 (123) and 33:08.09557 (124) pop defects 25 and equivalents, which fall within the scope of this invention. Although the image display has been described in terms of generating emulated images of analog audio discs, any data possessing an innate periodicity lends itself to this type of display. The effective rotational period of the display could be adapted to suit the periodicity of the available data. A record were located. The time offset between these defects is ing of an electrocardiogram of a human or animal is a suitable 1.78845 seconds. Comparing this to the defects used for the example of this sort of data. Presuming an average heart rate above calibration example, 26: 10.90964 (90) and 26:12.69802 (100) which are separated by 1.78838, the for a particular patient of 60 HZ, with a primary periodicity of roughly 1 HZ, a long time record of events could be displayed on the virtual platter surface. By setting the virtual display resultant percentage difference in rotational error between using these two measurements for calibration is 100* (1.78845—1.78838)/1.78838) or less than 0.004 percent dif ference. While it’s possible that the close agreement is fortu itous, more likely it indicates that the variation in turntable rotational velocity accuracy is probably small over the time needed to digitally record and transcribe an analog audio disc. 30 rotational rate at 2 HZ, 120 heartbeat events would be dis played per revolution. Each platter could show the equivalent of 30 minutes or more of the electrocardiogram recording. A steady heart rate would be re?ected by events aligned along 35 immediately apparent upon visual examination of the platter image. In contrast, discerning ?uctuations in data periodicity by the visual examination of a linear, orthogonal x-y plot FIG. 7B is a more detailed view of the result of the cali bration 109, where the overlap of the iconic representations of peak maxima demonstrate that the calibration successfully corrects the effective rotational rate. well de?ned radii, similar to the example for the calibrated disc rotational rate above. Any variations in rate would be 40 FIG. 7A shows the offset of the maximum iconically rep would be much more dif?cult over the time frame envisioned resented as 110 with time offset indicated 115. The next here. It should also be noted that there are many alternative ways revolution (arrow 114 indicates the direction of rotation) of the platter image brings maximum 111 into view, at time invention. For example, the virtual tone arm could be repre offset 116. Maximum 110 is offset because of rotational rate error to 110' relative to 111. A subsequent clockwise rotation of implementing the methods and apparatuses of the present 45 of the platter image brings maximum 112 into view at time offset 117. The previous maxima also are visible at 110' and 111'. Applying the calibration corrects the platter image for rotational rate error bringing the maxima into adjacent regis 50 as time offset 117 because maximum 112 was used as the point of reference for the rotational rate correction. According to an embodiment of the present invention, tools 55 rithms are sorted in tables according to the strength of the measured parameter. Two obvious methods consider the amplitude or slew rate of the signals as the pop detection parameter. The ?rst algorithm uses the maximum amplitude of the left or right signal channels. Of the 22 candidates displayed in the list 119, three (126, 127, 128) were physical pop events (con?rmed by examining the waveform and audi mental or marketing purposes, the emulated platter image may be impressed on the surface of the Disc, or used in the by analyZing the digitally sampled audio. This facilitates the calibration procedure described above. Three separate algorithms were considered for “pop” detection. In FIG. 8, analysis results using the three algo of the cylinder. The methods and apparatuses of the present invention may be used to generate an emulated platter image from the con tents of a digital recording that is intended to be mastered to a Compact Disc or Digital Versatile (Video) Disc. For oma tration 109 at time offset 118. The latter time offset is the same for transcribing audio disc recordings to a digital format, includes tools for locating physical “pop” defects on the discs sented as a linear carriage as depicted, or a pivoted linear or curved virtual tone arm. The platter metaphor also could be extended to other periodic implementations, such as a cylin der with the data image applied to the inner or outer surface, with the time dependent axis parallel to the axis of symmetry packaging or marketing materials of the Disc, providing a design that has added appeal because it would indicate the 60 65 actual characteristics of the information contained on the Disc. According to an embodiment of the present, invention, the methods and systems may be used to convert discretely sampled audio data, such as music into the circular display format indicated by FIGS. 2 and 3. A display of data obtains that emulates the appearance of a popular format for music dissemination, the vinyl (or formerly shellac) record. The general familiarity of the public with such records and their US 8,793,580 B2 17 18 associated playback equipment is an advantage, as most per sons already possess an intuitive grasp of the concept of the played. The operation mode selected is indicated in ?eld 81 vinyl LP disc. Further, the simulation goes beyond a purely cosmetic, stylized rendition of the appearance of a vinyl of the groove is located, the cueing button 17 (FIG. 2A), is engaged. The portion of the waveform displayed on the on FIG. 5, for mark-IN. When the central part of the quiet area screen to the left of the mark-in location is repeated in a record, because the appearance of the groove modulations re?ects the actual audio content of the recording. looping fashion, and played back audibly over the computer’ s speakers. The mark-in location may then be ?ne-tuned by gently rotating the platter until only lead-in noise is audible. If Selecting a track marking or cueing point by moving the “tonearm” and spinning the disc was the intuitive means employed by professional disc jockeys during the vinyl LP the mark-in location were moved to past the beginning of the audio, a small snippet of the audio may be heard. The rotation format era. The method and system convert the discretely sampled of the platter while listening to the loop and watching the data into a display that emulates the vinyl record format, or waveform provides the user with interactive feedback. This record image. The record image may optionally comprise permits rapidly selecting the mark-in location. The mark-in features of a conventional vinyl record and record player, location then is con?rmed. such as, for example, a tonearm/which may be used as a way Next, the track (or album side) mark-out optionally is to edit and play back digital audio ?les. Areas of low modu lation between tracks are easily selected by dropping the tonearm “stylus” on the “vinyl” surface. The track beginning/ selected. The procedure is similar to selecting the mark-in. Mark-OUT button 88 is engaged. The stylus is positioned at the end of the previous track (or album side). When the mark-out location is successfully located, only the noise of the lead-out of the previous track is heard. If the mark-out location is adjusted to a location before the end of the audio, end is then precisely located by “grabbing” and rotating the platter image. 20 An exemplary mode of operation is described in detail below. First, an audio ?le consisting of a single or multiple a snippet of the lead-out of the audio is heard. tracks is opened with the application software con?gured This editing procedure is invaluable when used in conjunc tion with making high-quality, accurate transcriptions of according to the present invention. The source of the audio ?le may be a transcription from a vinyl LP, a digital recording from another source (such as a cassette tape or live concert music recordings from a vinyl to a digital format. Compared to music recordings sourced from digital master recordings, recording), or a digital recording copied from a CD or other digital source. and distributed in a digital format, the modulation in between tracks of a transcribed vinyl disk does not drop to silence, 25 Next, the present invention automatically analyzes the audio data and generates a realistic, accurate image of a 30 single-track or multiple track vinyl record platter. An example of this step of the operation is shown in FIGS. 1 and 2, The ter, because one merely uses the waveform display to cut or select the tracks at obvious, digitally “silen ” locations. However, digital silence doesn’t exist in analog transcrip user may specify the color of the ‘vinyl’ substrate, in the same sense that commercially released records sometimes are pressed on colored or clear vinyl for cosmetic or promotional purposes. The user may also choose from the Acoustic or because of record vinyl surface noise. When editing purely digital recordings, locating track mark points is a trivial mat 35 tions of vinyl, so it’s impossible to establish accurate track mark points based only on the appearance of the waveform. For instance, a gradual song fade-out or fade-in may be heard Physical rendering options, depending on the rendering intent quite noticeably even in the presence of vinyl background or personal preference. The user also may specify the platter rendering format, corresponding to those typically encoun noise, which may obscure the music, viewed as the wave tered, such as, for example, RPM (331/3, 16, 45, 78, etc.) and image size (e.g., 7", 10", 12"). If digitally transcribing an form. However, setting track marks interactively using both 40 the waveform and audible feedback eliminates the possibility of inadvertently placing a track mark before the actual fade analog music disc, the selected format may be the same as the out or after an actual fade-in. The ease of use of the visual format of the source medium. representation generated according to the present invention The user may also specify that an image be superimposed on the substrate, and the music “grooves” drawn over the 45 allows the user to intuitively grab and spin the “platter” to further re?ne and accelerate the editing process. image with varying degrees of transparency. The image may The procedure of setting track, marks may be repeated for be a digital photograph, drawing or an abstract design, for each track. In the case of a multisided transcription of a vinyl example. record album, provision is made to specify the number of If the audio recording is sourced from an analog LP con sisting of multiple individual sides, or a CD transcription of a sides that are present in the recording. When the lead-out 50 recording originally released as a vinyl LP, the platter image mark of the last track on side one has been determined, the label area is clicked. The software program con?gured to created according to the present invention may consist of a implement the present invention interprets this as moving to single “side” comprised of all the tracks. Next, the toneam1/ stylus assembly is used to assign track the next side of the album. Track marks and song titles may continued to be added. This is repeated until all album sides are completed. Either before or after establishing track markers, the user mark points. The user may assign a track mark for each individual track, or only marks to delineate the sides of the LP record that is the source of the digital transcription. In the second instance, a two-sided transcription of a vinyl LP may be assigned four mark points. These mark points would cor respond to the music lead-in of side 1, the music lead-out of side 1, the music lead-in of side 2, and the music lead-out of 55 may optionally calibrate the accurate rotational velocity of the platter image (and putative playback speed) of the vinyl transcription. This calibration procedure depends on locating 60 side 2.As an alternative, the user could assign marks and titles to all individual tracks. aid in selecting suitable defects. For example, for a 33 1/3 RPM vinyl LP, at least one pair of defects must be located that are The procedure for setting the track marks is described in detail above with reference to FIG. 5. As described above, after moving the toneam1/ stylus to a blank modulation groove, the record is “spun” and the audio waveform dis physical defects in the audio recording caused by scratches or blemishes on the source disc. According to an embodiment of the present invention, one or more tools may be provided to 65 spaced approximately 1.8 seconds apart. The spacing depends on the putative rotational rate (1 6, 33 1/3, 45, 78 RPM, etc.) of the analog source disc.
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project