Electronic data processing apparatus and method for sound synthesis using transfer functions of sound samples

• US 6,208,969 B1
• Filed: 07/24/1998
• Issued: 03/27/2001
• Est. Priority Date: 07/24/1998
• Status: Expired due to Term

• Current Assignee: WSOU Investments, LLC (WSOU Holdings, LLC)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Curtin, Steven DeArmond
• Primary Examiner(s): Dorvil, Richemond
• Application Number:
  US09/122,520
• Time in Patent Office: 977 Days
• Field of Search: 704/258, 704/200, 704/201, 704/500, 704/212, 704/220, 704/203, 704/205, 704/206, 704/226, 704/227, 704/228, 704/260, 704/264, 704/265, 704/263
• US Class Current: 704/264
• CPC Class Codes: G10H 1/125 using a digital filter
• CPC Class Codes: G10H 2250/191 Chebyshev polynomials, e.g. to provide filter coefficients for sharp rolloff filters
• CPC Class Codes: G10H 2250/251 Wavelet transform, i.e. transform with both frequency and temporal resolution, e.g. for compression of percussion sounds; Discrete Wavelet Transform [DWT]
• CPC Class Codes: G10H 2250/625 Interwave interpolation, i.e. interpolating between two different waveforms, e.g. timbre or pitch or giving one waveform the shape of another while preserving its frequency or vice versa
• CPC Class Codes: G10H 7/00 Instruments in which the tones are synthesised from a data store, e.g. computer organs
• CPC Class Codes: G10L 13/04 Details of speech synthesis systems, e.g. synthesiser structure or memory management

Abstract

A method and an electronic data processing apparatus for wave synthesis that retains the true qualities of naturally occurring sounds, such as those of musical instruments, speech, or other sounds. Transfer functions representative of recorded sound samples are pre-calculated and stored for use in an interpolative process to generate a transfer function representative of the sound to be synthesized. The preferred transfer functions are Chebyshev polynomial-based transfer functions, which assure a highly predictable harmonic content of synthesized sound. Output sound generation is driven by time domain signals produced by reconversion of a sequence of interpolated transfer functions. Non-harmonic sounds are synthesized using multiple frequency inputs to the reconverting (waveshaping) stage, or by parallel waveshaping stages. Speech sibilants and noise envelopes of instruments are synthesized by the input of noise into the waveshaping stage by modulation of a sinusoid with band-limited noise.

Claims

1. A method of sound synthesis, comprising the steps of:

reading a frame of stored data that include transfer functions representing data derived from recorded sounds;

combining the transfer functions from the frame of stored data to effect spectral interpolation between harmonic data, yielding resultant transfer functions;

converting the resultant transfer functions to time domain signals; and

generating sounds from the time domain signals.

2. The method according to claim 1, wherein the reading, combining, and converting steps occur in a first process, the method further comprising in a second process conducted at least partially in parallel with the first process, the steps of:

reading a frame of stored data that include transfer function representing harmonic data derived from actual sounds;

combining transfer functions from the respective frames of stored data of the first and second processes to effect spectral interpolation between harmonic data represented in the respective first and second processes, yielding corresponding resultant transfer functions; and

converting the corresponding resultant transfer function to corresponding time domain signals,

whereby the sound generating step generates sound from the corresponding time domain signals.

3. The method according to claim 2, wherein the reading steps of the first and second processes read transfer functions representing harmonic data having differing timbre, whereby the sound generating step yields timbre morphing.

4. The method according to claim 3, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions having selected ranges of coefficients and representing recorded sounds of musical instruments having different timbres according to the ranges of coefficients.

5. The method according to claim 2, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions.

6. The method according to claim 2, wherein the converting step is driven at least in part by a plurality of waves that are not harmonically related.

7. The method according to claim 2, wherein the converting steps of the first and second processes are driven by respective waves that are modulated by respective band-limited noise signals.

8. The method according to claim 1, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions.

9. The method according to claim 1, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions having selected ranges of coefficients and representing recorded sounds of musical instruments having different timbres according to the ranges of coefficients.

10. The method according to claim 1, wherein the converting step includes modulating a band-limited noise signal on a sinusoidal excitation wave.

11. The method according to claim 1, wherein the stored transfer functions are represented by the coefficients of Chebyshev polynomials and wherein the reading step comprises the step of reading the coefficients into short-term memory as needed by the interpolation process and then evaluating the Chebyshev polynomials.

12. The method according to claim 11, wherein the step of reading the coefficients into short-term memory as needed comprises the step of reading the coefficients into short-term memory when an integer address value is changed.

13. The method according to claim 1, wherein the reading step comprises reading a frame of stored data that include transfer functions representing data derived from recorded sounds of a first musical instrument.

14. The method according to claim 13, wherein the step of converting the resultant transfer functions to time domain signals comprises the step filtering a waveform derived from a second musical instrument to a band close to its fundamental frequency and applying the filtered waveform to convert the resultant transfer functions interpolated from the transfer functions representing the data derived from recorded sounds of the first musical instrument.

15. The method according to claim 1, wherein the step of converting the resultant transfer functions to time domain signals comprises the step filtering a waveform derived from an external sound source to a band close to its fundamental frequency and applying the filtered waveform to convert the resultant transfer functions.

16. An electronic data processing system for additive sound synthesis, comprising:

an electronic memory storing a plurality of frames of data that include transfer functions representing harmonic data derived from recorded sounds;

a transfer function reader for reading from the memory a sequence of transfer functions;

apparatus for combining sequences of transfer functions to effect spectral interpolation between harmonic data, yielding resultant transfer functions;

excitation apparatus for converting the combined sequences of transfer functions to time domain signals; and

a speaker for generating sound from the time domain signals.

17. The electronic data processing system according to claim 16, wherein the transfer function reader comprises a portion of a first synthesis channel, the system further comprising a second synthesis channel in parallel with the first synthesis channel, the second synthesis channel including:

apparatus for reading sequences of transfer functions from the memory and furnishing them in the second channel,

whereby the apparatus for combining sequences of transfer functions effects spectral interpolation between harmonic data by interpolation between transfer functions respectively in the first and second synthesis channels, yielding corresponding resultant sequences of transfer functions input to the excitation apparatus.

18. The electronic data processing apparatus according to claim 17, wherein the transfer function reader and the apparatus for reading sequences of transfer functions from the memory and furnishing them in the second channel respectively read sequences of transfer functions representing harmonic data having differing timbre, whereby the sound generating step yields timbre morphing.

19. The electronic data processing apparatus according to claim 18, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions having selected ranges of coefficients and representing recorded sounds of musical instruments having different timbres according to the ranges of coefficients, and wherein the transfer function reader and the apparatus for reading sequences of transfer functions from the memory and furnishing them in the second channel respectively read sequences of transfer functions representing harmonic data having differing timbre, whereby the excitation apparatus and the speaker produce output sound having timbre morphing.

20. The electronic data processing apparatus according to claim 17, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions.

21. The electronic data processing apparatus according to claim 17, wherein the excitation apparatus includes means for generating waves that are not harmonically related.

22. The electronic data processing apparatus according to claim 17, wherein the excitation apparatus includes means for modulating at least one sinusoidal wave by a band-limited noise signal.

23. The electronic data processing apparatus according to claim 16, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions.

24. The electronic data processing apparatus according to claim 16, wherein the stored transfer functions include Chebyshev polynomial-based transfer functions having selected ranges of coefficients and representing recorded sounds of musical instruments having different timbres according to the ranges of coefficients.

25. The electronic data processing apparatus according to claim 16, wherein the excitation apparatus includes means for modulating a band-limited noise signal on a sinusoidal wave to produce excitation waves.

26. The electronic data processing apparatus according to claim 16, wherein the excitation apparatus includes means for generating a plurality of waves that, at least in part, are not harmonically related.

27. The electronic data processing system according to claim 16, wherein the excitation apparatus for converting the combined sequences of transfer functions to time domain signals comprises an analog apparatus.