Language-dependent state clustering for multilingual acoustic modelling

Niesler, Thomas

doi:10.1016/j.specom.2007.04.001

Cited by 31 publications

(25 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Usually, a language with large amounts of training data is used to simulate small amounts of target training data (Imseng et al, 2012a;Qian et al, 2011). For instance (Niesler, 2007) studied the sharing of resources on real underresourced languages, including Afrikaans, inspired by multilingual acoustic modeling techniques proposed by Schultz and Waibel (2001). However, only marginal ASR performance gains were reported.…”

Section: Introductionmentioning

confidence: 99%

Using out-of-language data to improve an under-resourced speech recognizer

Imseng

Motlíček

Bourlard

et al. 2014

Speech Communication

View full text Add to dashboard Cite

Under-resourced speech recognizers may benefit from data in languages other than the target language. In this paper, we report how to boost the performance of an Afrikaans automatic speech recognition system by using already available Dutch data. We successfully exploit available multilingual resources through (1) posterior features, estimated by multilayer perceptrons (MLP) and (2) subspace Gaussian mixture models (SGMMs). Both the MLPs and the SGMMs can be trained on out-of-language data. We use three different acoustic modeling techniques, namely Tandem, Kullback-Leibler divergence based HMMs (KL-HMM) as well as SGMMs and show that the proposed multilingual systems yield 12% relative improvement compared to a conventional monolingual HMM/GMM system only trained on Afrikaans. We also show that KL-HMMs are extremely powerful for under-resourced languages: using only six minutes of Afrikaans data (in combination with out-of-language data), KL-HMM yields about 30% relative improvement compared to conventional maximum likelihood linear regression and maximum a posteriori based acoustic model adaptation.

show abstract

Section: Introductionmentioning

confidence: 99%

Using out-of-language data to improve an under-resourced speech recognizer

Imseng

Motlíček

Bourlard

et al. 2014

Speech Communication

View full text Add to dashboard Cite

show abstract

“…However when multilingual ASR for many languages is desired, data collection and labeling may become too costly that alternative solutions are often desired. One potential solution is to explore shared acoustic phonetic structures among different languages to build a large set of acoustic models (e.g., [1][2] [3][5] [7][8] [9]) that characterize all the phone units needed in order to cover all the spoken languages being considered. This is sometimes called multilingual ASR or cross-lingual ASR when no language-specific data is available to build the acoustic models for the target language.…”

Section: Introductionmentioning

confidence: 99%

A study on multilingual acoustic modeling for large vocabulary ASR

Lin

Deng

et al. 2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

104

View full text Add to dashboard Cite

We study key issues related to multilingual acoustic modeling for automatic speech recognition (ASR) through a series of large-scale ASR experiments. Our study explores shared structures embedded in a large collection of speech data spanning over a number of spoken languages in order to establish a common set of universal phone models that can be used for large vocabulary ASR of all the languages seen or unseen during training. Language-universal and language-adaptive models are compared with language-specific models, and the comparison results show that in many cases it is possible to build general-purpose language-universal and language-adaptive acoustic models that outperform language-specific ones if the set of shared units, the structure of shared states, and the shared acoustic-phonetic properties among different languages can be properly utilized. Specifically, our results demonstrate that when the context coverage is poor in language-specific training, we can use one tenth of the adaptation data to achieve equivalent performance in cross-lingual speech recognition.

show abstract

“…Both Wheatley et al 1994 andSchultz andWaibel 2001 found that useful gains could be obtained by sharing data across languages with the size of the benefit dependent on the similarity of the sound systems of the languages combined. In the only crosslingual adaptation study using African languages (Niesler 2007), similar gains have not yet been observed.…”

Section: Asr For Resource-scarce Languagesmentioning

confidence: 88%

Collecting and evaluating speech recognition corpora for 11 South African languages

Badenhorst

Heerden

Davel

et al. 2011

Lang Resources & Evaluation

View full text Add to dashboard Cite

We describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which contains data from the eleven official languages of South Africa. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and we report on our investigation of the stability of the ASR models derived from the corpus. We also report on phoneme distance measures across languages, and describe initial phone recognisers that were developed using this data. We find that a surprisingly small number of speakers (fewer than 50) and around 10 to 20 h of speech per language are sufficient for the purposes of acceptable phone-based recognition.

show abstract

Language-dependent state clustering for multilingual acoustic modelling

Cited by 31 publications

References 14 publications

Using out-of-language data to improve an under-resourced speech recognizer

Using out-of-language data to improve an under-resourced speech recognizer

A study on multilingual acoustic modeling for large vocabulary ASR

Collecting and evaluating speech recognition corpora for 11 South African languages

Contact Info

Product

Resources

About