Presentation is loading. Please wait.

Presentation is loading. Please wait.

Separation of Multispeaker Speech Using Excitation Information B.Yegnanarayana, R.Kumara Swamy and S.R.Mahadeva Prasanna Dept of Computer Science and.

Similar presentations


Presentation on theme: "Separation of Multispeaker Speech Using Excitation Information B.Yegnanarayana, R.Kumara Swamy and S.R.Mahadeva Prasanna Dept of Computer Science and."— Presentation transcript:

1

2

3 Separation of Multispeaker Speech Using Excitation Information B.Yegnanarayana, R.Kumara Swamy and S.R.Mahadeva Prasanna Dept of Computer Science and Engineering Indian Institute of Technology Madras Chennai-600036, India yegna@cs.iitm.ernet.in Talk at NOLISP2005 April 19, 2005

4 Multispeaker Speech Signal Three speaker case ) Ta) Microphone-1 signal b) Microphone-2 signal

5 Multispeaker Whispered Speech Three Speaker case Ta) Microphone-1 signal b) Microphone-2 signal

6 Problem Determine the # speakers Separate individual speakers Enhance speech of individual speakers

7 Organization of the talk Demo illustrating the problem of multispeaker separation Basis: Sequences of impulses in speech production Proposed method for speaker separation Discussion: Scope of the present study and key ideas Conclusions

8 Basis for the Proposed Method of Separation Sequences of impulses in direct speech at mic locations No effect of channel or other degradations on the sequence No two speakers are at the same location

9 Proposed Method for Speaker Separation Record multispeaker data at 2 or more mics Compute the HE of the LP residual Use peaks in crosscorrelation of HEs to obtain delays Take min of shifted HEs to derive HE of desired speaker Derive weight function and modified LP residual Synthesize speech for each speaker

10 LP analysis of Speech signal Ta) Speech signal b) LP residual c) Hilbert Envelope of LP residual

11 Hilbert Envelope (HE) Ta) HE of microphone-1 signal b) HE of microphone-2 signal

12 Cross-Correlation of Hilbert Envelopes

13 Time-delay estimation (b) Time delay and normalized # samples (a) Peaks in the crosscorrelation plots

14 Processing HE using time-delay Ta) HE of mic-1 signal b), c), d) Min(HE1,HE2) emphasizing excitation information of Speaker 1,2 and 3, respectively

15 Results of Separation a)LP residual of mic-1 signal b), c) and d) modified residual of sp1, sp2 Sp3 e), f) and g) Speech signals after separation

16 Demo of Speaker Enhancement Three speaker case a a) Microphone-1 speech signal b) Microphone-2 speech signal ( a) b)

17 Demo of Speaker Enhancemnt a) Speaker 1 b) Speaker 2 c) Speaker 3

18 Summary Number of speakers (whispered), speaker separation (2 mics), speech enhancement (> 2 mics) Only speaker separation is addressed Significance of HE for delay estimation and speaker separation Conclusions Need to improve the quality of enhanced speech signals Need more microphones for data collection Need to deal with moving speaker and variable # speakers

19 Thank you very much for your attention


Download ppt "Separation of Multispeaker Speech Using Excitation Information B.Yegnanarayana, R.Kumara Swamy and S.R.Mahadeva Prasanna Dept of Computer Science and."

Similar presentations


Ads by Google