Previewing digital audio clips Number:7,149,593 from the United States Patent and Trademark Office (PTO) owispatent Essentially all of the processing parameters which control processing of a source audio signal to produce an encoded audio signal are stored in an audio processing profile. Multiple audio processing profiles are stored in a processing profile database such that specific combinations of processing parameters can be retrieved and used at a later time. Audio processing profiles are organized according to specific delivery bandwidths such that a sound engineer can quickly and efficiently encode audio signals for each of a number of distinct delivery media. Synchronized A/B switching during playback of various encoded audio signals allows the sound engineer to detect nuances in the sound characteristics of the various encoded audio signals.<H Previewing, digital, Previewing, digi, 7149593.html, Patent, pto, 7149593

Title: Previewing digital audio clips

Abstract: Essentially all of the processing parameters which control processing of a source audio signal to produce an encoded audio signal are stored in an audio processing profile. Multiple audio processing profiles are stored in a processing profile database such that specific combinations of processing parameters can be retrieved and used at a later time. Audio processing profiles are organized according to specific delivery bandwidths such that a sound engineer can quickly and efficiently encode audio signals for each of a number of distinct delivery media. Synchronized A/B switching during playback of various encoded audio signals allows the sound engineer to detect nuances in the sound characteristics of the various encoded audio signals.

Patent Number: 7,149,593 Issued on 12/12/2006 to Wiser, et al.

Inventors: Wiser; Philip R. (Redwood City, CA), Heringer; Leeann (Redwood City, CA), Kearby; Gerry (Redwood City, CA), Rishniw; Leon (Redwood City, CA), Brownell; Jason S. (Redwood City, CA)
Assignee: Microsoft Corporation (Redmond, WA)

Appl. No.: 11/171,467

Filed: June 29, 2005

Related U.S. Patent Documents


Application Number	Filing Date	Patent Number	Issue Date
10976393	Oct., 2004
08966072	Nov., 1997	6959220

Current U.S. Class: 700/94

Current International Class: G06F 17/00 (20060101)

Field of Search: 700/94 709/217,219

References Cited [Referenced By]

U.S. Patent Documents


5055939	October 1991	Karamon et al.
5418713	May 1995	Allen
5576843	November 1996	Cookson et al.
5636276	June 1997	Brugger
5734823	March 1998	Saigh et al.
5734891	March 1998	Saigh
5794217	August 1998	Allen
5805715	September 1998	Rhee
5845251	December 1998	Case
5895124	April 1999	Tsuga et al.
5928330	July 1999	Goetz et al.
5953506	September 1999	Kalra et al.
5996022	November 1999	Krueger et al.
6078669	June 2000	Maher
6185625	February 2001	Tso et al.
6263505	July 2001	Walker et al.
6341166	January 2002	Basel
6393198	May 2002	LaMacchia
6704421	March 2004	Kitamura
2003/0191548	October 2003	McPherson et al.
2005/0193094	September 2005	Robbin et al.

Other References

Digital Audio Compression Standard (AC-3), Dec. 20, 1995, Advanced Television Systems Committee, ATSC Standard. cited by other.

Primary Examiner: Pendleton; Brian
Attorney, Agent or Firm: Klarquist Sparkman, LLP

Parent Case Text

CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. application Ser. No. 10/976,393, filed Oct. 29, 2004, entitled "Digital Audio Signal Filtering Mechanism And Method", which is a continuation of U.S. application Ser. No. 08/966,072, filed Nov. 7, 1997, now U.S. Pat. No. 6,959,220, entitled "Digital Audio Signal Filtering Mechanism And Method", both of which are incorporated herein by reference.

Claims

What is claimed is:

1. A method for processing a source audio file to produce an encoded audio file, the encoded audio file corresponding in audio content to a portion of audio content of the source audio file for subsequent transmission through one of two or more transmission mediums selected for the transmission of the encoded audio file, each of the two or more transmission mediums having one or more transmission medium characteristics, the method comprising: storing an audio processing profile corresponding to each of the two or more transmission mediums, wherein the audio processing profile comprises processing parameter data for processing the source audio file in two or more cascaded processing stages, wherein the processing parameter data is based on the one or more of the transmission medium characteristics of the transmission mediums; associating data specifying the portion of the audio content of the source audio file to be encoded with at least some of the audio processing profiles corresponding to each of the two or more transmission mediums; receiving signals which specify one of the two or more transmission mediums selected for the transmission of the encoded audio file; retrieving the processing parameter data corresponding to the transmission medium selected for the transmission of the encoded audio file; and processing the specified portion of the audio content of the source audio file in the two or more cascaded processing stages in accordance with the retrieved processing parameter data to produce the encoded audio file for subsequent transmission through the transmission medium selected for the transmission of the encoded audio file.

2. The method of claim 1, further comprising transmitting the encoded audio file comprising encoded audio content corresponding to the portion of the audio content of the source audio file specified to be encoded upon receiving a request for transmission.

3. The method of claim 2, wherein the signals which specify the transmission medium selected for the transmission of the encoded audio file are received along with the request for transmission.

4. The method of claim 1, wherein the data specifying the portion of the audio content of the source audio file to be encoded comprises a start position and an end position corresponding to a first and last sample of the audio content of the source audio file to be encoded.

5. The method of claim 1, wherein the data specifying the portion of the audio content of the source audio file to be encoded further comprises a fade-in-field and a fade-out-field.

6. The method of claim 1, wherein the two or more cascaded processing stages further include a sample rate converter stage; (i) further wherein the processing parameter data for each of the two or more transmission mediums include delivery sample rate data which specify a transmission sample rate of the encoded audio file; (ii) further wherein the processing in the two or more cascaded processing stages further comprises re-sampling the audio content of source audio file in accordance with the sample rate data to produce a decimated/interpolated intermediate signal which has the transmission sample rate; and (iii) further wherein the two or more cascaded processing stages further comprises a step of filtering comprising filtering the decimated/interpolated intermediate signal in the filtering stage in accordance with filtering parameter data of the retrieved processing parameter data to produce a filtered audio signal.

7. The method of claim 1, wherein the two or more cascaded processing stages further include a format converter stage; (i) further wherein the processing parameter data for each of the two or more transmission mediums include transmission channel type data which specify a transmission channel type which is selected from the group consisting of a stereo channel type and mono-aural channel type; (ii) further wherein the processing in the two or more cascaded processing stages further comprises converting the audio content of the source audio file to the delivery channel type in accordance with the transmission channel type data to produce a format converted intermediate signal which has the transmission channel type; and (iii) further wherein the two or more cascaded processing stages further comprises a step of filtering comprising filtering the format converted intermediate signal in the filtering step in accordance with a filtering parameter data of the retrieved processing parameter data to produce a filtered audio signal.

8. A method for processing a source audio file to generate and transmit preview portions of a plurality of encoded versions of the source audio file, the preview portions corresponding in audio content to a portion of audio content of the source audio file, the method comprising: processing the source audio file according to a plurality of audio processing profiles to generate the plurality of encoded versions of the source audio file, the plurality of audio processing profiles corresponding to a plurality of transmission mediums on which the encoded versions of the source audio file can be transmitted, wherein the plurality of the audio processing profiles comprise processing parameter data for processing the source audio file in two or more cascaded processing stages, wherein the processing parameter data is based on one or more transmission medium characteristics of the plurality of transmission mediums; storing data specifying the preview portions of the plurality of the encoded versions of the source audio file; receiving signals which specify at least one of the plurality of transmission mediums selected for the transmission of the encoded version of the source audio file; and based on the signals specifying the at least one of the plurality of transmission mediums selected for the transmission and the data specifying the preview portions of the plurality of the encoded versions of the source audio file, retrieving at least one of the preview portions of the encoded versions of the source audio file and transmitting the at least one of the preview portions over the specified transmission medium.

9. The method of claim 8, further comprising transmitting the at least one of the preview portions over the specified transmission medium in response to a request.

10. The method of claim 9, wherein the request for transmission is sent along with the signals which specify at least one of the plurality of transmission mediums.

11. The method of claim 8, wherein the data specifying the preview portions of the plurality of the encoded versions of the source audio file comprises a starting position and an ending position for indicating a portion of at least one of the plurality of the encoded version of the source audio file.

12. The method of claim 11, wherein the data specifying the preview portions of the plurality of the encoded versions of the source audio file comprises a fade-in-field and a fade-out-field for fading in and fading out of a playback of the preview portions.

13. The method of claim 8, wherein the audio processing profiles further comprises a download enable flag field for indicating whether a corresponding one of the plurality of the encoded versions of the source audio file is permitted to be transmitted.

14. The method of claim 13 wherein the audio processing profiles further comprises a paid flag field for indicating whether a payment is required prior to a transmission of the corresponding one of the plurality of the encoded versions of the source audio file.

15. At least one computer-readable medium having stored thereon instructions for executing a method of processing a source audio file on a computer system having a processor and a memory, the method comprising: processing the source audio file according to a plurality of audio processing profiles to generate the plurality of encoded versions of the source audio file, the plurality of audio processing profiles corresponding to a plurality of transmission mediums on which the encoded versions of the source audio file can be transmitted, wherein the plurality of the audio processing profiles comprise processing parameter data for processing the source audio file in two or more cascaded processing stages, wherein the processing parameter data is based on one or more transmission medium characteristics of the plurality of transmission mediums; storing data specifying the preview portions of the plurality of the encoded versions of the source audio file; receiving signals which specify at least one of the plurality of transmission mediums selected for the transmission of the encoded version of the source audio file; and based on the signals specifying the at least one of the plurality of transmission mediums selected for the transmission and the data specifying the preview portions of the plurality of the encoded versions of the source audio file, retrieving at least one of the preview portions of the encoded versions of the source audio file and transmitting the at least one of the preview portions over the specified transmission medium.

16. The computer readable medium of claim 15, wherein the method of processing the source audio file further comprises transmitting the at least one of the preview portions over the specified transmission medium in response to a request.

17. The computer readable medium of claim 16, wherein the request for transmission is sent along with the signals which specify at least one of the plurality of transmission mediums.

18. The computer readable medium of claim 15, wherein the data specifying the preview portions of the plurality of the encoded versions of the source audio file comprises a starting position and an ending position for indicating a portion of at least one of the plurality of the encoded version of the source audio file.

19. The computer readable medium of claim 18, wherein the data specifying the preview portions of the plurality of the encoded versions of the source audio file comprises a fade-in-field and a fade-out-field for fading in and fading out of a playback of the preview portions.

20. The computer readable medium of claim 15, wherein the audio processing profiles further comprises a download enable flag field for indicating whether a corresponding one of the plurality of the encoded versions of the source audio file is permitted to be transmitted.

21. The computer readable medium of claim 20, wherein the audio processing profiles further comprises a paid flag field for indicating whether a payment is required prior to a transmission of the corresponding one of the plurality of the encoded versions of the source audio file.

Description

FIELD OF THE INVENTION

The present invention relates to computer filtering of digital audio signals and, in particular, to a particularly useful user interface for computer filtering of digital audio signals for delivery through various types of signal delivery media in a computer network.

BACKGROUND OF THE INVENTION

Audio signals stored in digital form have been in use for decades; however, distribution of such digital audio signals has generally been limited to physical distribution of tangible storage media in which the digital audio signals are encoded. Examples include compact discs (CDs) and digital audio tape (DAT) which store audio signals representing, for example, pre-recorded music, spoken word recordings, and sound effects. Recently, wide area computer networks such as the Internet have experienced tremendous growth in use and popularity. Accordingly, direct delivery of digital audio signals through such a wide area network has become an alternative to, and threatens to replace, physical delivery of tangible storage media as the primary delivery mode of digital audio signals.

Many digital audio signal filtering systems are currently available. Many such systems are used, for example, in producing a "master" signal in which various component signals, e.g., each from a separate musical instrument, are filtered and mixed such that the resulting master signal represents the artistic creation of an artist or collection of collaborating artists. This master signal is what is typically fixed in the tangible storage media which is physically distributed to the consuming public.

In direct delivery of digital audio signals through wide-area computer networks, the master signal can be sent directly to the computer system of a consumer. The master signal can be played directly from the consumer's computer system through a sound card and attached loudspeakers or can be stored on a tangible storage medium, e.g., writeable CD-ROM, for playback using conventional CD players and analog stereo equipment. Since the master signal is digital and is the same master signal which would traditionally be stored in tangible storage media by the producer, the master signal received by the consumer through the wide-area computer network is of the same quality as the master signal physically distributed on tangible storage media.

Sometimes, samples of the master signal are made available to the consumer through the computer network for preview purposes. Such samples are frequently streamed, i.e., delivered to a client computer system while the client computer system decodes and plays the received digital audio signals in real time. Because of variations in bandwidth with which various client computer systems are attached to computer networks such as the Internet, such samples are frequently delivered through low bandwidth communications media which are incapable of real-time delivery of such digital audio signals in a native, un-compressed form. Accordingly, the digital audio signal is generally compressed and encoded to reduce the amount of data required to represent the digital audio signal. The digital audio signal can be transmitted through computer network media in less time, requiring less bandwidth, than would ordinarily be required to transmit the digital audio signal in its native, un-encoded form. However, compression of the digital audio signal usually results in loss of detail of the digital audio signal such that sound quality of a received, decoded digital audio signal is typically degraded from the sound quality of the original digital audio signal prior to encoding and delivery through a computer network.

To mitigate the loss of signal quality as a result of such compression or to reduce some of the annoying effects of such compression, some sound engineers apply filters to a digital audio signal to enhance the result of compressing and encoding the digital audio signal. For example, in certain circumstances, emphasizing certain frequencies while de-emphasizing other frequencies of a digital audio signal prior to compression and encoding produces an encoded digital audio signal which has a more pleasant sound when decoded and played back relative to the sound of playback of a digital audio signal which is not filtered prior to such encoding. However, finding a particularly good combination of filters and encoders for a particular digital audio signal typically requires application of different filters from different suppliers and iterative application of such filters with various encoders to find an optimal combination. Furthermore, once a good combination of filters and encoders is determined for a particular digital audio signal, the combination is often not the best combination for a different digital audio signal and the entire empirical selection of a good combination of filters and encoders must generally be repeated for the different digital audio signal.

In addition, when distributing digital audio signals through a wide area computer network, it is sometimes desirable to deliver the digital audio signal within a particular amount of time. Such is desirable when streaming digital audio signals for real time playback. In such circumstances, the encoding of the digital audio signal should be tailored to the particular bandwidth of the network communications media connecting a particular recipient computer system with a source computer system within the computer network. In heterogeneous computer networks, various recipient computer systems can be connected with widely different bandwidths. For example, computer systems connected to the Internet are connected through network media ranging from 14.4 k modems to dedicated T1 connections which have many times the bandwidth of 14.4 k modems. Accordingly, encoding a digital audio signal for one recipient computer system having a particular bandwidth produces an encoded audio signal which is unacceptable for other recipient computer systems. For example, encoding a digital audio signal for real time delivery through a 14.4 k modem produces an encoded audio signal in which signal quality is unnecessarily sacrificed if the encoded signal is delivered to a recipient computer system connected to the source computer system through a dedicated T1 connection. Conversely, encoding a digital audio signal for real time delivery through a dedicated T1 connection produces an encoded audio signal which exceeds the available real-time delivery bandwidth of a recipient computer system connected to the source computer system through a 14.4k modem.

Further exacerbating the problem is that application of a combination of filters prior to encoding to produce a reasonably good quality encoded audio signal when encoded for a particular delivery bandwidth can produce an encoded audio signal of unacceptable quality when the same combination of filters is applied prior to encoding the digital audio signal for a different delivery bandwidth. Accordingly, a new combination of filters must be empirically determined for each delivery bandwidth for which a digital audio signal is to be encoded. Therefore, the amount of experimentation with various filters and encoders to deliver reasonably high quality signals to recipient computer systems connected through media of differing bandwidths can be overwhelming.

What is needed is a digital audio signal filtering and encoding system which significantly simplifies the processing of digital audio signals for distribution through heterogeneous computer networks through different delivery bandwidths.

SUMMARY OF THE INVENTION

In accordance with the present invention, an audio signal processor filters and encodes a source digital audio signal according to one or more audio processing profiles to produce one or more encoded audio signals. The audio signal processor includes an audio signal processing pipeline which performs all pre-processing, filtering, and encoding of the source audio signal to form each of the encoded audio signals. Each audio processing profile includes data specifying parameters of the pre-processing, filtering, and encoding. Thus, a single audio processing profile specifies all steps in processing the source audio signal to form an encoded audio signal.

The audio processing profiles for a particular source audio signal are organized according to specific delivery bandwidths. For example, one or more audio processing profiles are stored in a collection of audio processing profiles associated with a delivery bandwidth of 14.4 kbps. Other audio processing profiles are stored in collections of audio processing profiles associated with delivery bandwidths of 28.8 kbps, single-channel ISDN, dual-channel ISDN, and non-real time delivery. Non-real time delivery is generally not constrained by the bandwidth of a delivery medium. By organizing audio processing profiles according to associated delivery bandwidth, a sound engineer can empirically determine, and store for subsequent re-use, audio processing profiles which specify particularly good combinations of pre-processing, filtering, and encoding parameters for each of a number of different delivery bandwidths. For example, a sound engineer can determine that a specific combination of pre-processing, filtering, and encoding yields good results for real-time delivery of recordings of a string quartet through a delivery bandwidth of 14.4 kbps and that a different combination of pre-processing, filtering, and encoding yields good results for real-time delivery of recordings of a string quartet through a delivery bandwidth of 28.8 kbps. By storing such audio processing profiles, the sound engineer can quickly process, filter, and encode other recordings of string quartets easily and quickly over both 14.4 kbps and 28.8 kbps delivery media without requiring addition experimentation. A simple name, such as "Strings 14.4" or "Strings 28.8," can be used to identify the complex combination of processing parameters stored in such an audio processing profile and can be thereafter used by the sound engineer as a shorthand notation for that complex combination of processing parameters.

In addition, more than one audio processing profile can be created and stored for a particular delivery bandwidth. Thus, while the sound engineer previously empirically determined a particularly good combination of processing parameters for recordings of a string quartet, the sound engineer can also empirically determine a particular good combination of processing parameters for recordings of operas The sound engineer can therefore create a number of audio processing profiles for various respective types of sound recording for each delivery bandwidth and store such audio processing profiles for subsequent use to quickly and efficiently process additional digital audio signals of each various type with particularly good results. Furthermore, storage of such audio processing profiles allows a novice sound engineer to process audio signals using audio processing profiles created by other, more experienced sound engineers. In fact, a number of audio processing profiles can be directly programmed into an audio signal processor in accordance with the present invention and such audio processing profiles can therefore be made available to all sound engineers who use the audio signal processor.

Further in accordance with the present invention, the user can control A/B switching of playback of the source audio signal and one or more encoded audio signals. During playback of one audio signal, a graphical user interface receives signals from a user input device in response to physical manipulation by the user. In response thereto, the graphical user interface ceases playback of that audio signal and substantially immediately begins synchronized playback of another audio signal. The two signals can include the source audio signal and any of the encoded signals derived from the source audio signal and therefore have substantially the same sonic content, albeit pre-processed, filtered, and encoded. The playback of the second audio signal is synchronized in that the sonic content of the two audio signals seem uninterrupted to the user who hears the playback even though the general quality and tonal characteristics of the perceived sound will likely change when the playback switches from one to the other audio signal. As a result, a sound engineer can compare two different encoded audio signals formed by pre-processing, filtering, and encoding the source audio signal according to a slightly different set of process parameters stored in the audio processing profile. By switching between playback of the two encoded audio signals, the sound engineer can detect subtle differences in the quality of the sound of the two encoded audio signals and can make very fine adjustments in the processing, filtering, and encoding to achieve very good results. In addition, the sound engineer can compare the encoded audio signal which is the end result of such fine adjustments to the source audio signal using the same A/B switching technique to hear how the encoded audio signal compares to the source audio signal.

Inclusion of many parameters associated with various stages in the pre-processing, filtering and encoding of the source audio signal facilitates rapid iterative processing of digital audio signals to more quickly achieve a satisfactory combination of pre-processing, filtering, and encoding. Since all stages of audio signal processing, filtering, and encoding affect the quality of the end result, a comprehensive audio processing profile which controls the processing of every stage of the processing pipeline allows the user to change one or more parameters of one or more of the stages and to subsequently cause the processing pipeline to re-process each and every stage of the audio processing in accordance with new parameters. Thus, a sound engineer can manipulate every step of the processing from the source audio signal to the encoded audio signal in accordance with the changed parameters. Accordingly, the iterative process of empirically determining satisfactory combinations of processing, filtering and encoding parameters is accelerated.

These features combine to enable sound engineers to quickly and efficiently select combinations of pre-processing, filtering, and encoding parameters that yield particularly good results when encoding various types of digital audio signals for real-time delivery through a variety of delivery bandwidths.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of an audio signal processor in accordance with the present invention.

FIG. 2 is a block diagram of an audio profile database of the audio signal processor of FIG. 1.

FIG. 3 is a block diagram of a profile collection of the audio profile database of FIG. 2.

FIG. 4 is a block diagram of an audio processing profile of the profile collection of FIG. 3.

FIG. 5 is a block diagram of an equalization parameter field of the audio processing profile of FIG. 4.

FIG. 6 is a block diagram of a dynamic filtering parameter field of the audio processing profile of FIG. 4.

FIG. 7 is a block diagram of a computer system within which the audio signal processor of FIG. 1 executes.

FIG. 8 is a screen view of a preview pane of the audio signal processor of FIG. 1.

FIG. 9 is a block diagram of the source audio file processed by the audio signal processor of FIG. 1.

FIG. 10 is a block diagram of the header of the resulting composite resulting audio file of FIG. 1.

FIG. 11 is a pull-down menu by which a user selects one of a number of audio processing profiles according to the present invention.

FIG. 12 is a screen view of an audio processing profile edit window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 13 is a screen view of a sample rate window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 14 is a screen view of an equalizer window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 15 is a screen view of a dynamic filtering window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 16 is a screen view of a watermark window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 17 is a screen view of an encoder window by which the user can configure audio signal processing parameters in accordance with the present invention.

FIG. 18 is a screen view of a block diagram of an encoder parameter field of the audio processing profile of FIG. 4.

FIG. 19 is a screen view of a block diagram of a watermark parameter field of the audio processing profile of FIG. 4.

FIG. 20 is a second screen view of a preview pane of the audio signal processor of FIG. 1.

FIG. 21 is a block diagram of a mark field of the encoded audio signal of FIG. 10.

FIG. 22 is a block diagram of a fade field of the encoded audio signal of FIG. 10.

DETAILED DESCRIPTION

In accordance with the present invention, an audio signal processor 100 (FIG. 1) filters and encodes a source audio file 102 according to one or more audio processing profiles to produce a composite resulting audio file 104 which includes one or more encoded audio signals 106A E. Audio signal processor 100 includes a processing pipeline 114 which performs all pre-processing, filtering, and encoding of source audio file 102 to form composite resulting audio file 104. Specifically, processing pipeline 114 includes a format converter 116, a sample rate converter 118, an audio effects processor 120, a watermark processor 122, and an encoder 124, all of which are described more completely below. Briefly, (i) format converter 116 converts source audio file 102 from a stereo format to a single-channel format and vice versa and can change precision of each sample of source audio file 102; (ii) sample rate converter 118 converts a converted signal received from format converter 116 from the sampling rate of source audio file 102, e.g., 44 kHz, to a different sampling rate specified by a user; (iii) audio effects processor 120 performs various types of signal filtering on the re-sampled signal received from sample rate converter 118, such filtering including input gain, low shelf, bard pass, high shelf, expansion, compression, limiting, output gain, reverberation, and stereo imaging filtering; (iv) watermark processor 122 adds encoded identification data to the filtered signal received from signal processor 120; and (v) encoder 124 compresses the watermarked signal received from watermark processor 122 into a standardized format suitable for delivery through a computer network for subsequent decoding and playback.

Each of mono/stereo converter 116, sample rate converter 118, signal processor 120, watermark processor 122, and encoder 124 process received audio signals according to a number of parameters specified by a user. The particular parameters which, when used within processing pipeline 114, produce the best result in terms of sound quality and required delivery bandwidth of each of encoded audio signals 106A E depends upon the amount of delivery bandwidth available and the nature of the substantive sonic content of source audio file 102. Accordingly, such parameters are represented in a number of audio processing profiles which are stored in audio profile database 110. Audio profile database 110 is shown in greater detail in FIG. 2.

Audio profile database 110 includes a number of profile collections 202A E, each of which corresponds to a particular delivery bandwidth. For example, profile collections 202A, 202B, 202C, and 202D include audio processing profiles tailored by a user for use with delivery bandwidths of 14.4 kbps, 28.8 kbps, single-channel ISDN, and dual-channel ISDN connections. Profile collection 202E includes audio processing profiles tailored by a user for use with the delivery bandwidth of a T1 or better connection or for non-real-time download delivery in which delivery bandwidth is not a limiting concern. Profile collections 202A E are analogous to one another. Accordingly, the following description of profile collection 202A is equally applicable to profile collections 202B E.

Profile collection 202A is shown in greater detail in FIG. 3 and includes a number of audio processing profiles 304A E and a selector 302. Each of audio processing profiles 304A E specifies a number of processing parameters which control the processing of source audio file 102 (FIG. 1) by processing pipeline 114. The particular values of processing parameters stored in each of audio processing profiles 304A E can be selected for processing of a particular type of audio signal. For example, audio processing profile 304A can have processing parameter values selected for optimal processing of classical music such that as much of the balance and clarity of the audio signal is preserved during encoding. Audio processing profile 304B can have processing parameter values selected for optimal processing of grunge music characterized by heavily over-saturated electrical amplification of guitars. Others of audio processing profiles 304A E can have processing parameter values selected for optimal processing of other types of audio signals including spoken word, jazz music, nature recordings, and sounds in which preservation of stereo channels is more important that the quality of an equivalent single-channel audio signal. Selector 302 contains data which specifies one of audio processing profiles 304A E as the particular audio processing profile according to which source audio file 102 (FIG. 1) is to be processed for delivery through the delivery bandwidth with which profile collection 202B is associated, e.g., 28.8 kbps.

Audio processing profile 304A is shown in greater detail in FIG. 4. Audio processing profiles 304A E (FIG. 3) are analogous to one another. Accordingly, FIG. 4 and the following description of audio processing profile 304A are equally applicable to audio processing profiles 304B E.

Audio processing profile 304A includes a number of fields, each of which contain a collection of data defining a particular characteristic of audio processing profile 304A. Specifically, audio processing profile 304A includes a title field 400, a format field 402, a sample rate field 404A, a conversion quality field 404B, an equalization parameters field 406, a dynamic filtering parameters field 408, a watermark parameters field 410, and an encoder parameters field 412.

Title field 400 contains alphanumeric data by which a user identifies audio processing profile 304A. The particular data stored in title field 400 can be specified by the user by conventional user interface techniques and can be descriptive of the nature of processing specified by the parameters of audio processing profile 304A or of the type of audio signal yields particularly good results when processed according to audio processing profile 304A Format field 402 contains format data specifying an initial processing format of the audio signal of source audio file 102. Such format data includes (i) data specifying whether format converter 116 (FIG. 1) is to produce a mono-aural single-channel signal or a stereo, dual-channel signal from source audio file 102; (ii) data specifying a sample precision; and (iii) data specifying quantization compensation effects. When processing in accordance with audio processing profile 304A (FIG. 4), format converter 116 processes source audio file 102 in the following manner. If source audio file 102 stores a stereo signal and format field 402 contains data indicating that the resulting signal should be mono-aural, format converter 116 preforms a stereo-to-mono conversion on audio source file 102 to form a mono-aural intermediate signal. If source audio file 102 stores a mono-aural signal and format field 402 contains data indicating that the resulting signal should be stereo, format converter 116 performs a mono-to-stereo conversion on audio source file 102 to for a stereo intermediate signal. Otherwise, format converter 116 performs no conversion an the intermediate signal produced by format converter 116 is source audio file 102 in an unchanged form. Mono-to-stereo and stereo-to-mono conversions are well known. In one embodiment, stereo-to-mono conversion by format converter 116 is accomplished by reducing both channels in gain by 3 dB and summing the channels to produce a single, mono-aural channel.

In addition, format converter 116 converts the data word size, and precision, of each sample of the audio signal of source audio file 102 to a processing format as specified in format field 402 (FIG. 4). In addition, format converter 116 (FIG. 1) can dither the audio signal of source audio file 102 in accordance with dithering data stored in format field 402 (FIG. 4) to mask potentially undesirable effects of quantization in subsequent encoding. Dithering of digital audio signals is well-known.

Sample rate field 404A (FIG. 4) contains data specifying a delivery sample rate. When operating in accordance with audio processing profile 304A, sample rate converter 118 (FIG. 1) decimates or interpolates the intermediate signal produced by format converter 116 from the sample rate of source audio file 102 to the sample rate specified by the data contained in sample rate field 404A (FIG. 4). Decimation and interpolation of digital audio signals are well-known. In one embodiment, sample rate converter 118 (FIG. 1) uses a simple filter bank sample rate conversion technique based upon rational upsampling and downsampling ratios.

Sample rate converter 118 performs the rate conversion with a degree of signal quality specified by the data stored in sample rate field 404A. The result of processing by sample rate converter 118 is a decimated/interpolated intermediate signal.

Conversion quality field 404B (FIG. 4) contains data specifying a desired degree of fidelity of the decimated/interpolated intermediate signal to the original audio signal of source audio file 102 (FIG. 1). In one embodiment, the degree of fidelity is expressed as an interpolation filter size in which a high quality setting specifies a filter cut-off of -3 dB at a frequency of 90% of the delivery sample rate and in which a low quality setting specifies a filter cut-off of -3 dB at a frequency of 70% of the delivery sample rate. The larger filter of the higher quality setting reduces the amount of noise and distortion of the decimated/interpolated intermediate signal relative to the original audio signal of source audio file 102.

Equalization parameters field 406 (FIG. 4) contains data specifying parameters of an input gain filter and a number of low-shelf, band-pass, and high-shelf filters applied to the decimated/interpolated intermediate signal by audio effects processor 120 (FIG. 1) to produce an equalized intermediate signal. Equalization parameters field 406 (FIG. 4) is shown in greater detail in FIG. 5. Equalization parameters field 406 includes an input gain field 502 and fields specifying four separate filters. Each of the four filters includes a type field, a frequency field, a gain field, and a Q field such as type field 504A, frequency field 506A, gain field 508A, and Q field 510A, for example. Input gain field 502 contains data specifying an amount of gain to add to the decimated/interpolated intermediate signal prior to filtering according to fields 504A D, 506A D, 508A D, and 510A D. Each of type fields 504A D contains data specifying one of three types of filters, namely, a low shelf filter, a band-pass filter, or a high shelf filter, for each respective one of the four filters. Each of frequency fields 506A''D contains data specifying a corresponding filter frequency for each respective one of the four filters. For band-pass filters, the specified filter frequency is the center of the frequency band of the filter. For low shelf filters, the specified filter frequency is the upper limit of the filtered frequencies. For high shelf filters, the specified filter frequency is the lower limit of the filtered frequencies. Each of gain fields 508A D contains data specifying a corresponding gain for each respective one of the four filters. Each of Q fields 510A D contains data specifying the filter selectivity for each respective one of the four filters. The filter selectivity, e.g., as represented in Q field 510B, is expressed as a ratio of the center frequency, e.g., as represented in frequency field 506B, to a passband. For example, if the center frequency represented in frequency field 506B is 1 kHz and Q field 510B indicates a filter selectivity of 5, the passband for the associated filter is 200 Hz. In this illustrative embodiment, the passband is measured at the points at which the gain is -3 dB. Data in Q fields 510A D are ignored for respective ones of the four filters which are not band-pass filters as indicated by respective ones of type fields 504A D.

Processing by audio effects processor 120 (FIG. 1) in accordance with equalization parameters field 406 (FIG. 5) includes adjusting the decimated/interpolated intermediate signal by an amount of gain specified by data contained in input gain field 502 and processing the adjusted signal by the four filters specified in fields 504A D, 506A D, 508A D, and 510A D. The resulting signal is an equalized intermediate signal.

When operating in accordance with audio processing profile 304A (FIG. 4), audio effects processor 120 (FIG. 1) further filters the equalized intermediate signal in accordance with parameters represented in dynamic filtering parameters field 408 (FIG. 4) which is shown in greater detail in FIG. 6. Dynamic filtering parameters field 408 includes a bypass field 602, a stereo link field 604, an attack time field 606, a release time field 608, an expander ratio field 610, an expander threshold field 612, a compressor ratio field 614, a compressor threshold field 616, a limiter threshold field 618, an output gain field 620, and an output gain makeup field 622. Bypass field 602 contains data indicating whether filtering according to data stored in dynamic parameters field 408 (FIG. 4) is to be performed or bypassed altogether. Stereo link field 604 (FIG. 6) contains data indicating whether the left and right channels of a stereo signal should be linked. If the left and right channels are linked, the same amount of gain is applied to both channels. Otherwise, different amounts of gain can be applied to each channel.

Attack time field 606 and release time field 608 contain data representing attack time and release time, respectively, of a gain profile applied to the equalized intermediate signal by audio effects processor 120 (FIG. 1).

Expander ratio field 610 (FIG. 6) contains data specifying a first order gain profile to be applied by audio effects processor 120 (FIG. 1) to the equalized intermediate signal below an amplitude specified by data stored in expander threshold field 612 (FIG. 6).

Compressor ratio field 614 contains data specifying a first order gain profile to be applied by audio effects processor 120 (FIG. 1) to the equalized intermediate signal above an amplitude specified by data stored in compressor threshold field 616 (FIG. 6) and below an amplitude specified by data stored in limiter threshold field 618. The amplitude specified in limiter threshold field 618, which is sometimes referred to as the limiter threshold, represents a clip amplitude such that any samples of the equalized intermediate signal having an amplitude over the limiter threshold are clipped by audio effects processor 120 (FIG. 1) to have the limiter threshold as their amplitude.

Output gain field 620 (FIG. 6) contains data specifying a fixed gain to be applied by audio effects processor 120 (FIG. 1) to the equalized intermediate signal. Processing the equalized intermediate signal by audio effects processor 120 (FIG. 1) in accordance with watermark parameters field 410 (FIG. 4) of audio processing profile 304A produces a filtered intermediate signal which is processed by watermark processor 122 (FIG. 1). Watermark processor 122 embeds identification data in the filtered intermediate signal such that subsequent decoding of the subsequently encoded audio signal can identify audio signal processor 100 as the source of the encoded audio signal. In one embodiment, watermark processor 122 is the Electronic DNA watermarking system available from Solana Technology Development Corporation of San Diego, Calif. Watermark processor 122 modulates a noise sequence dependent upon the filtered intermediate signal using the identification data and adds the modulated noise sequence to the filtered intermediate signal to thereby embed the identification data in the filtered intermediate signal. The identification data can later be extracted using conventional techniques. Watermark parameters field 410 (FIG. 19) includes a bypass field 1902, a quality field 1904, and a fidelity field 1906. Bypass field 1902 contains data indicating whether the filtered intermediate signal is to be watermarked at all. Quality field 1904 contains data specifying a degree of watermarking quality in terms of a level of robustness required for anticipated noise and distortion in delivery media and subsequent playback environments. Fidelity field 1906 contains data which specifies a degree of audibility of the watermark in the resulting watermarked intermediate signal. Processing of the filtered intermediate signal by watermark processor 122 in accordance with audio processing profile 304A (FIG. 4) produces a watermarked intermediate signal which is processed by encoder 124.

Encoder 124 processes the watermarked intermediate signal according to encoding parameters stored in encoding parameters field 412 (FIG. 4) to produce an encoded audio signal. Encoding parameters field 412 is shown in greater detail in FIG. 18 and includes a data rate field 1802, a compression field 1804, an optimization field 1806, a quality field 1808, an auxiliary data rate field 1810, a bandwidth field 1812, a channel coupling field 1814, a coupling frequency field 1816, a verbose mode field 1818, a bandwidth filter field 1820, a LFE filter field 1822, a LFE channel field 1824, a DC filter field 1820, a phase shift field 1828, and a de-emphasis field 1830. In one embodiment, encoder 124 (FIG. 1) is the AC-3 audio encoder available from Dolby Laboratories Inc. of San Francisco, Calif. In this illustrative embodiment, the fields of encoding parameters field 412 (FIG. 18) and their use is defined by the AC-3 audio encoder. A few of the fields of encoding parameters field 412 are described herein for completeness.

Data rate field 1802 contains data specifying the data rate, and thus the size, of the encoded audio signal. Optimization field 1806 stores data indicating whether audio quality it to be optimized for downloading audio. Audio can be downloaded by a customer on certain conditions described more completely below. Quality field 1806 stores data representing a desired degree of signal quality to be maintained during encoding of the filtered intermediate signal. Bandwidth filter field 1820 contains data specifying whether a low pass filter is applied prior to encoding the filtered intermediate signal. Bandwidth field 1812 contains data specifying a threshold frequency for the low pass filter. Channel coupling field 1814 contains data specifying whether left and right channels of the filtered intermediate signal are to be coupled during encoding at frequencies about a threshold frequency represented by data stored in coupling frequency field 1816. DC filter field 1826 contains data specifying whether a high pass filter is applied prior to encoding the filtered intermediate signal.

Encoder 124 (FIG. 1) encodes the filtered intermediate signal in accordance with encoding parameters 412 (FIG. 18) in the manner described above. The result of encoding the watermarked intermediate signal by encoder 124 is an encoded audio signal, i.e., one of encoded audio signals 106A E of composite resulting audio file 104.

Thus, data stored in audio processing profile 304A (FIG. 4) specifies characteristics of many steps of signal processing by which source audio file 102 is transformed into one of encoded audio signals 106A E, e.g., encoded audio signal 106A. Such characteristics include characteristics of stereo/mono-aural conversion, sample interpolation/decimation, various types of filtering, watermark processing, and encoding. The power and advantage of storing all such processing characteristics in a single audio processing profile such as audio processing profile 304A (FIG. 4) is more fully appreciated in the context of graphical user interface 112 (FIG. 1).

Graphical User Interface 112

Graphical User Interface (GUI) 112 facilitates user control of processing by audio signal processor 100. Specifically, GUI 112 receives user-generated signals responsive to physical manipulation by the user of one or more user input devices 730 (FIG. 7) of a computer system 700 within which audio signal processor 100 executes. Full appreciation of the present invention and of GUI 112 (FIG. 1) is facilitated by a more complete understanding of computer system 700 (FIG. 7), i.e., the operating environment of audio signal processor 100.

Computer system 700 (FIG. 7) includes a processor 702 and memory 704 which is coupled to processor 702 through an interconnect 706. Interconnect 706 can include generally any interconnect mechanism for computer system components and can include, e.g., a bus, a crossbar, a mesh, a torus, or a hypercube. Processor 702 fetches from memory 704 computer instructions and executes the fetched computer instructions. In addition, processor 702 can fetch computer instructions through a computer network 770 through network access circuitry 760 such as a POTS or ISDN modem or ethernet network access circuitry. Processor 702 also reads data from and writes data to memory 704 and sends data and control signals through interconnect 706 to one or more computer display devices 720 and receives data and control signals through interconnect 706 from one or more computer user input devices 730 in accordance with fetched and executed computer instructions.

Memory 704 can include any type of computer memory and can include, without limitation, randomly accessible memory (RAM), read-only memory (ROM), and storage devices which include storage media such as magnetic and/or optical disks. Memory 704 includes audio signal processor 100 which is all or part of a computer process which in turn executes within processor 702 from memory 704. A computer process is generally a collection of computer instructions and data which collectively define a task performed by a computer system such as computer system 700.

Each of computer display devices 720 can be any type of computer display device including without limitation a printer, a cathode ray tube (CRT), a light-emitting diode (LED) display, or a liquid crystal display (LCD). Each of computer display devices 720 receives from processor 702 control signals and data and, in response to such control signals, displays the received data. Computer display devices 720, and the control thereof by processor 702, are conventional.

Each of user input devices 730 can be any type of user input device including, without limitation, a keyboard, a numeric keypad, or a pointing device such as an electronic mouse, trackball, lightpen, touch-sensitive pad, digitizing tablet, thumb wheels, joystick, or voice recognition circuitry. Each of user input devices 730 generates signals in response to physical manipulation by a user and transmits those signals through interconnect 706 to processor 702.

Computer system 700 also includes audio processing circuitry 780 coupled to interconnect 706 and one or more loudspeakers 790 coupled to audio processing circuitry 780. Audio processing circuitry 780 receives audio signals and control signals from processor 702 through interconnect 706 and, in response thereto, produces sounds through loudspeakers 790. Since the user of audio signal processor 100 selects filtering parameters in a manner described more completely below based upon subtle nuances in the tonal qualities of filtered audio signals as played through audio processing circuitry 780 and loudspeakers 790, it is preferred that audio processing circuitry 780 and loudspeakers 790 are of relatively high quality and perform with relatively high fidelity. In one embodiment, audio processing circuitry 780 is the AudioMedia III sound card available from DigiDesign Inc. of Palo Alto, Calif. and loudspeakers 790 are the 20--20bas powered loudspeakers available from Event Electronics of Santa Barbara, Calif.

In one embodiment, computer system 700 is a computer system which is compatible with the PC personal computer available from International Business Machines, Inc. of Somers, N.Y., and processor 702 is based upon the architecture of the Pentium series of microprocessors available from Intel Corporation of Santa Clara, Calif. In this illustrative embodiment, computer system 700 executes, and operates under control of, the Microsoft Windows 95 operating system available from Microsoft Corporation of Redmond, Wash.

As described above, audio signal processor 100 executes within processor 702 from memory 704. Specifically, processor 702 fetches computer instructions from audio signal processor 100 and executes those computer instructions. Processor 702, in executing audio signal processor 100, reads digital audio signals from source audio file 102, processes and encodes those digital audio signals in the manner described above to form encode audio signals 106A E (FIG. 1) of composite resulting audio file 104.

GUI 112 of audio signal processor 100 presents the user of audio signal processor 100 with a particularly effective tool for selecting a combination of values for the audio signal processing parameters described about of an audio processing profile relatively quickly with relatively little effort. Two considerations regarding measuring relative quality of audio signals are central to the design of GUI 112. First, each step of processing between source audio file 102 and encoded audio signals 106A E affects the quality of encoded audio signals 106A E when decoded and played through audio processing circuitry 780 and loudspeakers 790