DocumentCode :
3510632
Title :
Towards source-filter based single sensor speech separation
Author :
Stark, Michael ; Pernkopf, Franz
Author_Institution :
Signal Process. & Speech Commun. Lab., Graz Univ. of Technol., Graz
fYear :
2009
fDate :
19-24 April 2009
Firstpage :
97
Lastpage :
100
Abstract :
We present a new source-filter based method to separate two speakers talking simultaneously at equal level mixed into a single sensor. First, the relation between the spectral whitened mixture and the speakers excitation signals is analyzed. Therefore, a factorial HMM capturing also time dependencies is exploited. Then, the estimated excitation signals are combined with best fitting vocal tract information taken from a trained dictionary. We report results on the database of Cooke considering 108 speech mixtures. The average improvement of 2.9 dB in SIR for all data is lower but not significantly lower compared to the Gaussian mixture method which relies on known pitch-tracks. Although the performance is currently moderate we believe in this approach and its significance towards the development of speaker independent single sensor speech separation.
Keywords :
Markov processes; source separation; speaker recognition; Gaussian mixture; best fitting vocal tract information; hidden Markow model; single sensor speech separation; source filter; speakers excitation signals; spectral whitened mixture; Background noise; Filters; Hidden Markov models; Laboratories; Linear predictive coding; Oral communication; Robust stability; Signal analysis; Signal processing; Speech processing; Hidden Markov Model; Source-Filter Representation; Speech Separation; Vector-Quantization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on
Conference_Location :
Taipei
ISSN :
1520-6149
Print_ISBN :
978-1-4244-2353-8
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2009.4959529
Filename :
4959529
Link To Document :
بازگشت