Phoneme Recognition on the TIMIT Database

Lopes, Carla; Perdigão, Fernando

doi:10.5772/17600

Cited by 68 publications

(50 citation statements)

References 36 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It would appear that higher results will not be achieved for the TIMIT corpus unless other approaches are used, which should include a widening of the phonetic context as was successfully done in [6]. Milestones in phone recognition using the TIMIT database can be found in [30].…”

Section: Contextualizing With Timit State-of-art Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Broad phonetic class definition driven by phone confusions

Lopes

Perdigão

2012

EURASIP J. Adv. Signal Process.

Self Cite

View full text Add to dashboard Cite

Intermediate representations between the speech signal and phones may be used to improve discrimination among phones that are often confused. These representations are usually found according to broad phonetic classes, which are defined by a phonetician. This article proposes an alternative data-driven method to generate these classes. Phone confusion information from the analysis of the output of a phone recognition system is used to find clusters at high risk of mutual confusion. A metric is defined to compute the distance between phones. The results, using TIMIT data, show that the proposed confusion-driven phone clustering method is an attractive alternative to the approaches based on human knowledge. A hierarchical classification structure to improve phone recognition is also proposed using a discriminative weight training method. Experiments show improvements in phone recognition on the TIMIT database compared to a baseline system.

show abstract

Section: Contextualizing With Timit State-of-art Resultsmentioning

confidence: 99%

“…This database is often used in phone recognition benchmarking, e.g. [5,7,30]. The train and test sets correspond to the original splitting of the TIMIT database.…”

Section: Resultsmentioning

confidence: 99%

Broad phonetic class definition driven by phone confusions

Lopes

Perdigão

2012

EURASIP J. Adv. Signal Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The "TIMIT recipe" contained in the Kaldi distribution 2 reproduces exactly the same evaluation settings described in (Lopes and Perdigao, 2011) for a phone recognition task based on this corpus. Moreover, Kaldi provides some n-best rescoring scripts that apply RNNLM hypothesis rescoring and interpolate the results with the standard N-gram model results used in the evaluation.…”

Section: Extrinsic Evaluationmentioning

confidence: 99%

“…In the speech community, the TIMIT corpus is the base for a standard phone-recognition task with specific evaluation procedures described in detail in (Lopes and Perdigao, 2011). We stick completely to this evaluation to test the effectiveness of our proposed model adopting, among the other procedures, the same splitting between the different data sets: the training set contains 3696 utterances (140225 phones), the validation set 400 utterances (15057 phones) and the test set 192 utterances (7215 phones).…”

Section: Datamentioning

confidence: 99%

Towards Quantum Language Models

Basile¹,

Tamburini²

2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

This paper presents a new approach for building Language Models using the Quantum Probability Theory, a Quantum Language Model (QLM). It mainly shows that relying on this probability calculus it is possible to build stochastic models able to benefit from quantum correlations due to interference and entanglement. We extensively tested our approach showing its superior performances, both in terms of model perplexity and inserting it into an automatic speech recognition evaluation setting, when compared with state-of-theart language modelling techniques.

show abstract

“…The cross-language forced alignment system uses a phone recogniser with a 21.5% phone error rate on the TIMIT corpus, so it is still fairly close to state-of-the-art (Schwarz, 2009, p46;Lopes and Perdigao, 2011). The artificial neural network uses a 310 ms window so it is implicitly context dependent (Schwarz, 2009, p39).…”

Section: Experimental Set-upmentioning

confidence: 99%

Cross-language forced alignment to assist community-based linguistics for low resource languages

Kempton¹

2017

Proceedings of the 2nd Workshop on the Use of Computational Methods in the Study of Endangered Languages

View full text Add to dashboard Cite

In community-based linguistics, community members become involved in the analysis of their own language. This insider perspective can radically increase the speed and accuracy of phonological analysis, e.g. providing rapid identification of phonemic contrasts. However, due to the nature of these community-based sessions, much of the phonetic data is left undocumented. Rather than going back to traditional fieldwork, this paper argues that corpus phonetics can be applied to recordings of the community-based analysis sessions. As a first step in this direction, cross-language forced alignment is applied to the type of data generated by a community-based session in the Nikyob language of Nigeria. The alignments are accurate and suggest that corpus phonetics could complement community-based linguistics giving community members a powerful tool to analyse their own language.

show abstract

Phoneme Recognition on the TIMIT Database

Cited by 68 publications

References 36 publications

Broad phonetic class definition driven by phone confusions

Broad phonetic class definition driven by phone confusions

Towards Quantum Language Models

Cross-language forced alignment to assist community-based linguistics for low resource languages

Contact Info

Product

Resources

About