Signal detection apparatus including maximum likelihood estimation and noise suppression

• US 5,355,431 A
• Filed: 11/27/1992
• Issued: 10/11/1994
• Est. Priority Date: 05/28/1990
• Status: Expired due to Term

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Kane, Joji, Nohara, Akira
• Primary Examiner(s): Knepper, David D.
• Application Number:
  US07/982,771
• Time in Patent Office: 683 Days
• Field of Search: 395/2, 395/2.35-2.37, 395/2.55, 381/46, 381/47, 381/94, 381/42
• US Class Current: 704/226
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 2021/02087 the noise being separate speech, e.g. cocktail party
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0272 Voice signal separating
• CPC Class Codes: G10L 25/12 the extracted parameters being prediction coefficients
• CPC Class Codes: G10L 25/24 the extracted parameters being the cepstrum
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10

Abstract

A band division process including a Fourier transformation is performed for an inputted speech signal, thereby outputting spectrum signals of plural channels. A cepstrum analysis process is performed for the spectrum signals, and a peak of the obtained cepstrum is detected in response to the cepstrum analysis result. Thereafter, a speech signal interval of the inputted noisy speech signal is detected in response to the detected peak, and a noise is predicted in the speech signal in response to the detected speech signal interval. Then, the predicted noise is canceled in the spectrum signals thereby outputting noise-suppressed spectrum signals. Finally, the noise-suppressed spectrum signals are combined and are inverse Fourier-transformed, thereby outputting a noise-suppressed speech signal.

Claims

1. A signal detection and noise suppression apparatus comprising:

storage means for storing information of predetermined features of a plurality of kinds of predetermined signals;

band division means for performing a band division process including a Fourier transformation for an inputted noisy speech signal and outputting spectrum signals of a plurality of channels;

cepstrum analysis means for performing a cepstrum analysis process for the spectrum signals of a plurality of channels outputted from said band division means and outputting a cepstrum analysis result;

peak detection means for detecting a peak of a cepstrum in response to the cepstrum analysis result outputted from said cepstrum analysis means;

speech judgment means for detecting a speech signal interval of the inputted noisy speech signal in response to the peak detected by said peak detection means and outputting the detected speech signal interval;

maximum likelihood estimation means for detecting a kind of the signal of the inputted noisy speech signal by comparing a feature of the inputted noisy speech signal with the features stored in said storage means;

noise prediction means for predicting a noise in the speech of the inputted noisy speech signal in response to both of the detected speech signal interval outputted from said speech judgment means and the kind of the signal of the inputted noisy speech signal detected by said maximum likelihood estimation means;

cancellation means for cancelling the noise predicted by said noise prediction means in response to the spectrum signals of a plurality of channels outputted from said band division means and outputting noise-suppressed spectrum signals of a plurality of channels; and

band combining means for combining the noise-suppressed spectrum signals of a plurality of channels, inverse Fourier-transforming the combined spectrum signal, and outputting a transformed speech signal.