Adaptation of speech recognition vocabularies for improved transcription of YouTube videos

Denis Jouvet (1) , David Langlois (2) , Mohamed Amine Menacer (2) , Dominique Fohr (1) , Odile Mella (1) , Kamel Smaïli (2)

1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
2 SMarT - Statistical Machine Translation and Speech Modelization and Text

This paper discusses the adaptation of speech recognition vocabularies for automatic speech transcription. The context is the transcription of YouTube videos in French, English and Arabic. Base-line automatic speech recognition systems have been developed using previously available data. However, the available text data, including the GigaWord corpora from LDC, are getting quite old with respect to recent YouTube videos that are to be transcribed. After a discussion on the performance of the ASR baseline systems, the paper presents the collection of recent textual data from internet for updating the speech recognition vocabularies and for training the language models, as well as the elaboration of development data sets necessary for the vocabulary selection process. The paper also compares the coverage of the training data collected from internet, and of the GigaWord data, with finite size vocabularies made of the most frequent words. Finally, the paper presents and discusses the amount of out-of-vocabulary word occurrences, before and after the update of the speech recognition vocabularies, for the three languages. Moreover, some speech recognition evaluation results are provided and analyzed.

Domaines: Informatique et langage [cs.CL]

Denis Jouvet, David Langlois, Mohamed Amine Menacer, Dominique Fohr, Odile Mella, et al.. Adaptation of speech recognition vocabularies for improved transcription of YouTube videos. Journal of International Science and General Applications, 2018, 1 (1), pp.1-9. ⟨hal-01873801⟩