Method and apparatus for separating sound-source signal and method and device for detecting pitch

• US 20050195990A1
• Filed: 02/17/2005
• Published: 09/08/2005
• Est. Priority Date: 02/20/2004
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Kondo, Tetsujiro, Ichiki, Hiroshi, Shima, Junichi, Arimitsu, Akihiko
• Granted Patent: US 8,073,145 B2
• Time in Patent Office: Days
• US Class Current: 381/92
• CPC Class Codes: G10L 21/0272 Voice signal separating
• CPC Class Codes: G10L 21/028 using properties of sound source
• CPC Class Codes: G10L 25/90 Pitch determination of speech signals

Abstract

In a sound-source signal separating method, a target sound-source signal in an input audio signal is enhanced, the input audio signal being from a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup devices. The pitch of the target sound-source signal in the input audio signal is detected, and the target sound-source signal is separated from the input audio signal based on the detected pitch and the enhanced sound-source signal.

Claims

1. A sound-source signal separating apparatus, comprising:

a sound-source signal enhancing unit operable to enhance a target sound-source signal in an input audio signal to produce an enhanced sound-source signal, the input audio signal including a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup devices;

a pitch detector operable to detect a pitch of the target sound-source signal in the input audio signal; and

a sound-source signal separating unit operable to separate the target sound-source signal from the input audio signal based on the detected pitch and the enhanced sound-source signal.

2. The sound-source signal separating apparatus according to claim 1, wherein the sound-source signal separating unit comprises:

a filter for separating the target sound-source signal from a signal output from the sound-source signal enhancing unit; and

a filter coefficient output unit operable to output a filter coefficient of the filter based on information detected by the pitch detector.

3. The sound-source signal separating apparatus according to claim 2, wherein the filter coefficient features a frequency characteristic of the filter which causes a frequency component to pass through the filter, the frequency component having a frequency which is an integer multiple of the frequency of the pitch of the target sound-source signal.

4. The sound-source signal separating apparatus according to claim 3, wherein the filter coefficient output unit comprises a memory storing filter coefficients corresponding to a plurality of pitches, the filter coefficient output unit reading and outputting a filter coefficient from the memory corresponding to the pitch of the target sound-source signal.

5. The sound-source signal separating apparatus according to claim 2, further comprising:

a high-frequency region processing unit operable to process a portion of the output signal in a consonant band; and

a filter bank operable to extract the portion of the output signal in the consonant band from the sound-source signal enhancing unit and to transfer the portion of the output signal in the consonant band to the high-frequency region processing unit, to extract a portion of the output signal in a band other than the consonant band from the sound-source signal enhancing unit and to transfer the portion of the output sound-source signal in the band other than the consonant band to the filter, and to extract a portion of the output signal in a vowel band from the sound-source signal enhancing unit and to transfer the portion of the output signal in the vowel band to the pitch detector.

6. The sound-source signal separating apparatus according to claim 2, wherein the plurality of sound pickup devices comprise a left stereomicrophone and a right stereomicrophone.

7. The sound-source signal separating apparatus according to claim 2, wherein the sound-source signal enhancing unit corrects the acoustic signals from the plurality of sound pickup devices with a time difference between sound propagation delays, each sound propagation delay being measured from a target sound source to each of the plurality of sound pickup devices, and adds the corrected acoustic signals from the plurality of sound pickup devices in order to enhance the acoustic signal from only the target sound source.

8. The sound-source signal separating apparatus according to claim 2, wherein the pitch detector detects the pitch of the target sound-source signal using two wavelengths of the pitch of the target sound-source signal as a unit of detection.

9. The sound-source signal separating apparatus according to claim 1, wherein the sound-source signal separating unit comprises:

a fundamental waveform producing unit operable to produce a fundamental waveform based on information detected by the pitch detector using a steady portion of a signal output from the sound-source signal enhancing unit, the steady portion having at least about the same pitch consecutively repeated throughout; and

a fundamental waveform substituting unit operable to substitute a repetition of the fundamental waveform produced by the fundamental waveform producing unit for at least a portion of a signal based on the input audio signal.

10. The sound-source signal separating apparatus according to claim 9, wherein the pitch detector detects the pitch of the target sound-source signal using two wavelengths of the pitch of the target sound-source signal as a unit of detection.

11. The sound-source signal separating apparatus according to claim 9, wherein the plurality of sound pickup devices comprise a left stereomicrophone and a right stereomicrophone.

12. The sound-source signal separating apparatus according to claim 9, wherein the sound-source signal enhancing unit corrects the acoustic signals from the plurality of sound pickup devices with a time difference between sound propagation delays, each sound propagation delay being measured from a target sound source to each of the plurality of sound pickup devices, and adds the corrected acoustic signals from the plurality of sound pickup devices in order to enhance the acoustic signal from only the target sound source.

13. The sound-source signal separating apparatus according to claim 9, wherein the fundamental waveform producing unit averages the target sound-source signal in a steady portion of the target sound-source signal having at least about the same pitch consecutively repeated throughout using two wavelengths of the pitch of the target sound-source signal as a unit of detection.

14. A sound-source signal separating method, comprising:

enhancing a target sound-source signal in an input audio signal to produce an enhanced sound-source signal, the input audio signal including a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup devices;

detecting a pitch of the target sound-source signal in the input audio signal; and

separating the target sound-source signal from the input audio signal based on the detected pitch and the enhanced sound-source signal.

15. A sound-source signal separating apparatus, comprising:

sound-source signal enhancing means for enhancing a target sound-source signal in an input audio signal to produce an enhanced sound-source signal, the input audio signal including a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup units;

pitch detector means for detecting a pitch of the target sound-source signal in the input audio signal; and

sound-source signal separating means for separating the target sound-source signal from the input audio signal based on the detected pitch and the enhanced sound-source signal.

16. A pitch detector, comprising:

a sound-source signal enhancing unit operable to enhance a target sound-source signal in an input audio signal to produce an enhanced sound-source signal, the input audio signal including a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup devices;

a period detector operable to detect a two-wavelength period of a signal output from the sound-source signal enhancing unit using two wavelengths of a pitch of the output signal as a unit of detection; and

a continuity determining unit operable to determine, in response to a change in the two-wavelength period detected by the period detector, whether at least about the same pitch has been consecutively repeated, and to output pitch information as the result of the determination.

17. The pitch detector according to claim 16, wherein the plurality of sound pickup devices comprise a left stereomicrophone and a right stereomicrophone.

18. The pitch detector according to claim 16, wherein the sound-source signal enhancing unit corrects the acoustic signals from the plurality of sound pickup devices with a time difference between sound propagation delays, each sound propagation delay being measured from a target sound source to each of the plurality of sound pickup devices, and adds the corrected acoustic signals from the plurality of sound pickup devices in order to enhance the acoustic signal from only the target sound source.

19. A pitch detecting method, comprising:

enhancing a target sound-source signal in an input audio signal to produce an enhanced sound-source signal, the input audio signal including a mixture of acoustic signals from a plurality of sound sources picked up by a plurality of sound pickup devices;

detecting a two-wavelength period of a signal output from the sound-source signal enhancing step using two wavelengths of a pitch of the output signal as a unit of detection; and

determining, in response to a change in the two-wavelength period detected in the period detecting step, whether at least about the same pitch has been consecutively repeated, and outputting pitch information as the result of the determination.

20. A sound-source signal separating apparatus, comprising:

a pitch detector operable to detect a pitch of a target sound-source signal of an input audio signal using a wavelength twice the pitch of the target sound-source signal as a unit of detection, the input audio signal including a mixture of acoustic signals from a plurality of sound sources; and

a sound-source signal separating unit operable to separate the target sound-source signal based on the detected pitch.

21. A sound-source signal separating method, comprising:

detecting a pitch of a target sound-source signal of an input audio signal using a wavelength twice the pitch of the target sound-source signal as a unit of detection, the input audio signal including a mixture of acoustic signals from a plurality of sound sources; and

separating the target sound-source signal based on the detected pitch.