分享

Introduced by Wang et al. in VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation

Introduced by Li et al. in Spoken SQuAD: A Study of Mitigating the Impact of Speech Recognition Errors on Listening Comprehension

The Switchboard-1 Telephone Speech Corpus (LDC97S62) consists of approximately 260 hours of speech and was originally collected by Texas Instruments in 1990-1,

Introduced by Boito et al. in MaSS: A Large and Clean Multilingual Corpus of Sentence-aligned Spoken Utterances Extracted from the Bible

Introduced by Wang et al. in CoVoST: A Diverse Multilingual Speech-To-Text Translation Corpus

Introduced by Fan et al. in CN-CELEB: a challenging Chinese speaker recognition dataset

Introduced by Ephrat et al. in Looking to Listen at the Cocktail Party: A Speaker-Independent Audio-Visual Model for Speech Separation

Introduced by Maciejewski et al. in WHAMR!: Noisy and Reverberant Single-Channel Speech Separation

The TIMIT Acoustic-Phonetic Continuous Speech Corpus is a standard dataset used for evaluation of automatic speech recognition systems.

Introduced by Iranzo-Sánchez et al. in Europarl-ST: A Multilingual Corpus For Speech Translation Of Parliamentary Debates

Introduced by Cosentino et al. in LibriMix: An Open-Source Dataset for Generalizable Speech Separation

Introduced by Ryant et al. in The Second DIHARD Diarization Challenge: Dataset, task, and baselines