Method, preprocessor, speech recognition system, and program product for extracting target speech by removing noise
View Patent ↗The present invention relates to a method, preprocessor, speech recognition system, and program product for extracting a target speech by removing noise. In an embodiment of the invention target speech is extracted from two input speeches, which are obtained through at least two speech input devices installed in different places in a space, applies a spectrum subtraction process by using a noise power spectrum (Uω) estimated by one or both of the two speech input devices (Xω(T)) and an arbitrary subtraction constant (α) to obtain a resultant subtracted power spectrum (Yω(T)). The invention further applies a gain control based on the two speech input devices to the resultant subtracted power spectrum to obtain a gain-controlled power spectrum (Dω(T)). The invention further applies a flooring process to said resultant gain-controlled power spectrum on the basis of arbitrary Flooring factor (β) to obtain a power spectrum for speech recognition (Zω(T)).
1. A method for extracting a target speech from two input speeches, wherein said input speeches are obtained using at least two speech input devices, said method comprising:
determining a noise power spectrum, wherein said determination uses at least one of said two input speeches;
applying a spectrum subtraction process to subtract at least part of said noise power spectrum from a speech spectrum of at least one of said two input speeches to obtain a subtracted power spectrum;
computing cross-power spectrum phase (CSP) coefficients based on said two input speeches, said CSP coefficients indicative of a direction of a speaker relative to one of said at least two speech input devices;
applying a gain control to said subtracted power spectrum based on said CSP coefficients, wherein said gain control amplifies speech received from a direction indicated by said CSP coefficients, to obtain a resultant power spectrum; and
applying a flooring process to the resultant power spectrum to obtain a processed power spectrum of the target speech suitable for speech recognition.
2. The method of claim 1 , further comprising:
applying an array process on said two input speeches to enhance sound received from a direction indicated by said CSP coefficients, and applying said spectrum subtraction process to the result of said array process.
3. The method of claim 2 , wherein said array process is a delay-sum array process using said two input speeches.
4. The method of claim 2 , wherein said array process is an adaptive array process using said two input speeches.
5. The method of claim 1 , wherein said at least two speech input devices are inside of a car.
6. The method of claim 1 , wherein at least one of said at least two speech input devices includes a non-directional microphone.
7. The method of claim 1 , wherein at least one of said at least two speech input devices includes a directional microphone.
8. The method of claim 1 , wherein computing CSP coefficients comprises computing an angle indicative of a direction of said speaker relative to one of the at least two speech input devices.
9. The method of claim 1 , further comprising:
applying a recognizer to extract at least one feature value from the processed power spectrum for use in obtaining a speech recognition result.
10. The method of claim 1 , wherein applying a flooring process to said resultant power spectrum comprises increasing at least one value of at least part of said resultant power spectrum.
11. A speech recognition system comprising:
a preprocessor for performing preprocessing to extract a target speech from two input speeches, wherein said input speeches are obtained using at least two speech input devices, said preprocessor comprising:
a noise determination process section configured to determine a noise power spectrum, wherein said determination uses at least one of said two input speeches;
a spectrum subtraction process section configured to apply a spectrum subtraction process to subtract at least part of said noise power spectrum from a speech spectrum of at least one of said two input speeches to obtain a subtracted power spectrum;
a directivity computation section configured to compute cross-power spectrum phase (CSP) coefficients based on said two input speeches, said CSP coefficients indicative of a direction of a speaker relative to one of said at least two speech input devices; and
a gain control section configured to apply gain control to said subtracted power spectrum based on said CSP coefficients, wherein said gain control amplifies speech received from a direction indicated by said CSP coefficients, to obtain a resultant power spectrum; and
a flooring process section configured to apply a flooring process to the resultant power spectrum to obtain a processed power spectrum of the target speech suitable for speech recognition.
12. The system of claim 11 , further comprising:
a recognizer configured to extract at least one feature value for use in obtaining a speech recognition result.
13. The system of claim 11 , wherein the flooring process section is configured to increase at least one value of at least part of said resultant power spectrum.
14. At least one computer-readable storage device storing computer-executable instructions that, when executed by at least one processor, perform a method for extracting a target speech from two input speeches, wherein said input speeches are obtained using at least two speech input devices, said method comprising:
determining a noise power spectrum, wherein said determination uses at least one of said two input speeches;
applying a spectrum subtraction process to subtract at least part of said noise power spectrum from a speech spectrum of at least one of said two input speeches to obtain a subtracted power spectrum;
computing cross-power spectrum phase (CSP) coefficients based on said two input speeches, said CSP coefficients indicative of a direction of a speaker relative to one of said at least two speech input devices;
applying a gain control to said subtracted power spectrum based on said CSP coefficients, wherein said gain control amplifies speech received from a direction indicated by said CSP coefficients, to obtain a resultant power spectrum; and
applying a flooring process to the resultant power spectrum to obtain a processed power spectrum of the target speech suitable for speech recognition.
15. The at least one computer-readable storage device of claim 14 , further comprising:
applying a recognizer to extract at least one feature value for use in obtaining a speech recognition result.
16. The at least one computer-readable storage device of claim 14 , wherein applying a flooring process to said resultant power spectrum comprises increasing at least one value of at least part of said resultant power spectrum.