The AMIDA 2009 Meeting Transcription System
| Type of publication: | Conference paper |
| Citation: | Hain_INTERSPEECH_2010 |
| Booktitle: | Proceedings of Interspeech |
| Year: | 2010 |
| Month: | 9 |
| Location: | Makuhari, Japan |
| Abstract: | We present the AMIDA 2009 system for participation in the NIST RT’2009 STT evaluations. Systems for close-talking, far field and speaker attributed STT conditions are described. Im- provements to our previous systems are: segmentation and diar- isation; stacked bottle-neck posterior feature extraction; fMPE training of acoustic models; adaptation on complete meetings; improvements to WFST decoding; automatic optimisation of decoders and system graphs. Overall these changes gave a 6- 13% relative reduction in word error rate while at the same time reducing the real-time factor by a factor of five and using con- siderably less data for acoustic model training. |
| Keywords: | |
| Projects |
AMIDA |
| Authors | |
| Added by: | [UNK] |
| Total mark: | 0 |
|
Attachments
|
|
|
|
|
Notes
|
|
|
|
|
