Výsledky vyhledávání

Report

Non-autoregressive Error Correction for CTC-based ASR with Phone-conditioned Masked LM

Autor: Futami, Hayato, Inaguma, Hirofumi, Ueno, Sei, Mimura, Masato, Sakai, Shinsuke, Kawahara, Tatsuya

Connectionist temporal classification (CTC) -based models are attractive in automatic speech recognition (ASR) because of their non-autoregressive nature. To take advantage of text-only data, language model (LM) integration approaches such as rescori

Externí odkaz: http://arxiv.org/abs/2209.04062

Zobrazit plný text záznamu

Report

Distilling the Knowledge of BERT for Sequence-to-Sequence ASR

Autor: Futami, Hayato, Inaguma, Hirofumi, Ueno, Sei, Mimura, Masato, Sakai, Shinsuke, Kawahara, Tatsuya

Attention-based sequence-to-sequence (seq2seq) models have achieved promising results in automatic speech recognition (ASR). However, as these models decode in a left-to-right way, they do not have access to context on the right. We leverage both lef

Externí odkaz: http://arxiv.org/abs/2008.03822

Zobrazit plný text záznamu

Report

End-to-end speech-to-dialog-act recognition

Autor: Dang, Viet-Trung, Zhao, Tianyu, Ueno, Sei, Inaguma, Hirofumi, Kawahara, Tatsuya

Spoken language understanding, which extracts intents and/or semantic concepts in utterances, is conventionally formulated as a post-processing of automatic speech recognition. It is usually trained with oracle transcripts, but needs to deal with err

Externí odkaz: http://arxiv.org/abs/2004.11419

Zobrazit plný text záznamu

Report

Speech Corpus of Ainu Folklore and End-to-end Speech Recognition for Ainu Language

Autor: Matsuura, Kohei, Ueno, Sei, Mimura, Masato, Sakai, Shinsuke, Kawahara, Tatsuya

Ainu is an unwritten language that has been spoken by Ainu people who are one of the ethnic groups in Japan. It is recognized as critically endangered by UNESCO and archiving and documentation of its language heritage is of paramount importance. Alth

Externí odkaz: http://arxiv.org/abs/2002.06675

Zobrazit plný text záznamu