Acoustic Modeling and Training of a Bilingual ASR System when a Minority Language is Involved
نویسندگان
چکیده
This paper describes our work in developing a bilingual speech recognition system using two SpeechDat databases. The bilingual aspect of this work is of particular importance in the Galician region of Spain where both languages Galician and Spanish coexist and one of the languages, the Galician one, is a minority language. Based on a global Spanish-Galician phoneme set we built a bilingual speech recognition system which can handle both languages: Spanish and Galician. The recognizer makes use of context dependent acoustic models based on continuous density hidden Markov models. The system has been evaluated on a isolated-word large-vocabulary task. The tests show that Spanish system exhibits a better performance than the Galician system due to its better training. The bilingual system provides an equivalent performance to that achieved by the language specific systems.
منابع مشابه
Development of bilingual ASR system for MediaParl corpus
The development of an Automatic Speech Recognition (ASR) system for the bilingual MediaParl corpus is challenging for several reasons: (1) reverberant recordings, (2) accented speech, and (3) no prior information about the language. In that context, we employ frequency domain linear prediction-based (FDLP) features to reduce the effect of reverberation, exploit bilingual deep neural networks ap...
متن کاملAcoustic and lexical resource constrained ASR using language-independent acoustic model and language-dependent probabilistic lexical model
One of the key challenges involved in building statistical automatic speech recognition (ASR) systems is modeling the relationship between subword units or “lexical units” and acoustic feature observations. To model this relationship two types of resources are needed, namely, acoustic resources i.e., speech data with word level transcriptions and lexical resources where each word is transcribed...
متن کاملUnsupervised acoustic model training using multiple seed ASR systems
Unsupervised acoustic modeling can offer a cost and time effective way of creating a solid acoustic model for any under-resourced language. This paper explores the novel idea of using two independent ASR systems to transcribe new speech data, align and filter the ASR hypotheses and use the presumably correct transcriptions to iteratively improve the two seed ASR systems. In parallel, the newly ...
متن کاملExploiting foreign resources for DNN-based ASR
Manual transcription of audio databases for the development of automatic speech recognition (ASR) systems is a costly and time-consuming process. In the context of deriving acoustic models adapted to a specific application, or in low-resource scenarios, it is therefore essential to explore alternatives capable of improving speech recognition results. In this paper, we investigate the relevance ...
متن کاملGrapheme-based Automatic Speech Recognition using Probabilistic Lexical Modeling
Automatic speech recognition (ASR) systems incorporate expert knowledge of language or the linguistic expertise through the use of phone pronunciation lexicon (or dictionary) where each word is associated with a sequence of phones. The creation of phone pronunciation lexicon for a new language or domain is costly as it requires linguistic expertise, and includes time and money. In this thesis, ...
متن کامل