DocumentCode :
2178006
Title :
Joint encoding of the waveform and speech recognition features using a transform codec
Author :
Fan, Xing ; Seltzer, Michael L. ; Droppo, Jasha ; Malvar, Henrique S. ; Acero, Alex
Author_Institution :
Microsoft Res., Redmond, WA, USA
fYear :
2011
fDate :
22-27 May 2011
Firstpage :
5148
Lastpage :
5151
Abstract :
We propose a new transform speech codec that jointly encodes a wideband waveform and its corresponding wideband and narrowband speech recognition features. For distributed speech recognition, wideband features are compressed and transmitted as side information. The waveform is then encoded in a manner that exploits the information already captured by the speech features. Narrowband speech acoustic features can be synthesized at the server by applying a transformation to the decoded wideband features. An evaluation conducted on an in-car speech recognition task show that at 16 kbps our new system typically shows essentially no impact in word error rate compared to uncompressed audio, whereas the standard transform codec produces up to a 20% increase in word error rate. In addition, good quality speech is obtained for playback and transcription, with PESQ scores ranging from 3.2 to 3.4.
Keywords :
speech codecs; speech recognition; speech synthesis; PESQ; in-car speech recognition; joint encoding; narrowband speech recognition features; speech acoustic features task; speech codec; transform codec; wideband speech recognition features; Codecs; Encoding; Mel frequency cepstral coefficient; Narrowband; Speech; Speech recognition; Wideband; Siren codec; distributed speech recognition; speech coding; transform coding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
Conference_Location :
Prague
ISSN :
1520-6149
Print_ISBN :
978-1-4577-0538-0
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2011.5947516
Filename :
5947516
Link To Document :
بازگشت