SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION
This paper presents the results of the study on improving robustness to the acoustic variability of the speech signal for spontaneous speech recognition system. The method is proposed to constructing high-level bottleneck features using deep neural network adapted to a speaker and to acoustic enviro...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)
2016-01-01
|
| Series: | Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki |
| Subjects: | |
| Online Access: | http://ntv.ifmo.ru/file/article/14589.pdf |
| _version_ | 1818207635016515584 |
|---|---|
| author | I. P. Medennikov |
| author_facet | I. P. Medennikov |
| author_sort | I. P. Medennikov |
| collection | DOAJ |
| description | This paper presents the results of the study on improving robustness to the acoustic variability of the speech signal for spontaneous speech recognition system. The method is proposed to constructing high-level bottleneck features using deep neural network adapted to a speaker and to acoustic environment with i-vectors. The proposed method provides 11,9% relative reduction of word error rate in Russian spontaneous telephone speech recognition task. |
| first_indexed | 2024-12-12T04:32:03Z |
| format | Article |
| id | doaj.art-e7b30b1107b443fc9ff51ab14c033e57 |
| institution | Directory Open Access Journal |
| issn | 2226-1494 2500-0373 |
| language | English |
| last_indexed | 2024-12-12T04:32:03Z |
| publishDate | 2016-01-01 |
| publisher | Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University) |
| record_format | Article |
| series | Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki |
| spelling | doaj.art-e7b30b1107b443fc9ff51ab14c033e572022-12-22T00:38:04ZengSaint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University)Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki2226-14942500-03732016-01-0116119519710.17586/2226-1494-2016-16-1-195-197SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITIONI. P. MedennikovThis paper presents the results of the study on improving robustness to the acoustic variability of the speech signal for spontaneous speech recognition system. The method is proposed to constructing high-level bottleneck features using deep neural network adapted to a speaker and to acoustic environment with i-vectors. The proposed method provides 11,9% relative reduction of word error rate in Russian spontaneous telephone speech recognition task.http://ntv.ifmo.ru/file/article/14589.pdfautomatic speech recognitionspeaker adaptationi-vectorsbottleneck features from deep neural network |
| spellingShingle | I. P. Medennikov SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki automatic speech recognition speaker adaptation i-vectors bottleneck features from deep neural network |
| title | SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION |
| title_full | SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION |
| title_fullStr | SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION |
| title_full_unstemmed | SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION |
| title_short | SPEAKER-DEPENDENT FEATURES FOR SPONTANEOUS SPEECH RECOGNITION |
| title_sort | speaker dependent features for spontaneous speech recognition |
| topic | automatic speech recognition speaker adaptation i-vectors bottleneck features from deep neural network |
| url | http://ntv.ifmo.ru/file/article/14589.pdf |
| work_keys_str_mv | AT ipmedennikov speakerdependentfeaturesforspontaneousspeechrecognition |