1 |
Unsupervised speech representation learning using WaveNet autoencoders ...
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Using Web Co-occurrence Statistics for Improving Image Categorization ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Phoneme and sentence-level ensembles for speech recognition
|
|
|
|
In: European Association for Speech, Signal and Image Processing. EURASIP Journal on audio, speech, and music processing. - Heidelberg : Springer (2011), 17 S.
|
|
UB Frankfurt Linguistik
|
|
Show details
|
|
4 |
Investigating Lexical Substitution Scoring for Subtitle Generation
|
|
|
|
In: http://infoscience.epfl.ch/record/146027 (2010)
|
|
BASE
|
|
Show details
|
|
5 |
Using more informative posterior probabilities for speech recognition
|
|
|
|
In: http://infoscience.epfl.ch/record/146087 (2010)
|
|
BASE
|
|
Show details
|
|
6 |
Investigating Lexical Substitution Scoring for Subtitle Generation
|
|
|
|
In: http://infoscience.epfl.ch/record/146409 (2010)
|
|
BASE
|
|
Show details
|
|
7 |
Using more informative posterior probabilities for speech recognition
|
|
|
|
In: http://infoscience.epfl.ch/record/146434 (2010)
|
|
Abstract:
In this paper, we present initial investigations towards boosting posterior probability based speech recognition systems by estimating more informative posteriors taking into account acoustic context (e.g., the whole utterance), as well as possible prior information (such as phonetic and lexical knowledge). These posteriors are estimated based on HMM state posterior probability definition (typically used in standard HMMs training). This approach provides a new, principled, theoretical framework for hierarchical estimation/use of more informative posteriors integrating appropriate context and prior knowledge. In the present work, we used the resulting posteriors as local scores for decoding. On the OGI numbers database, this resulted in significant performance improvement, compared to using MLP estimated posteriors for decoding (hybrid HMM/ANN approach) for clean and more specially for noisy speech. The system is also shown to be much less sensitive to tuning factors (such as phone deletion penalty, language model scaling) compared to the standard HMM/ANN and HMM/GMM systems, thus practically it does not need to be tuned to achieve the best possible performance.
|
|
URL: http://infoscience.epfl.ch/record/146434/files/ketabdar-idiap-rr-05-91.pdf http://publications.idiap.ch/downloads/reports/2005/ketabdar-idiap-rr-05-91.pdf http://infoscience.epfl.ch/record/146434
|
|
BASE
|
|
Hide details
|
|
8 |
Discriminative Keyword Spotting
|
|
|
|
In: http://infoscience.epfl.ch/record/146043 (2010)
|
|
BASE
|
|
Show details
|
|
14 |
Offline Recognition of Large Vocabulary Cursive Handwritten Text
|
|
|
|
In: http://infoscience.epfl.ch/record/82926 (2006)
|
|
BASE
|
|
Show details
|
|
15 |
Offline Recognition of Large Vocabulary Cursive Handwritten Text
|
|
|
|
In: http://infoscience.epfl.ch/record/82912 (2006)
|
|
BASE
|
|
Show details
|
|
16 |
Can a Professional Imitator Fool a GMM-Based Speaker Verification System?
|
|
|
|
In: http://infoscience.epfl.ch/record/83202 (2006)
|
|
BASE
|
|
Show details
|
|
17 |
Offline Recognition of Unconstrained Handwritten Texts Using HMMs and Statistical Language Models
|
|
|
|
In: http://infoscience.epfl.ch/record/83055 (2006)
|
|
BASE
|
|
Show details
|
|
18 |
Offline Recognition of Unconstrained Handwritten Texts Using HMMs and Statistical Language Models
|
|
|
|
In: http://infoscience.epfl.ch/record/82961 (2006)
|
|
BASE
|
|
Show details
|
|
19 |
An Investigation of Spectral Subband Centroids for Speaker Authentication
|
|
|
|
In: http://infoscience.epfl.ch/record/82970 (2006)
|
|
BASE
|
|
Show details
|
|
20 |
Evaluation of Formant-Like Features for ASR
|
|
|
|
In: http://infoscience.epfl.ch/record/82782 (2006)
|
|
BASE
|
|
Show details
|
|
|
|