Search Results for author:
Found 3 papers, 3 papers with code
GigaST: A 10,000-hour Pseudo Speech Translation Corpus
The training set is translated by a strong machine translation system and the test set is translated by human.
WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Research
To address this data scarcity issue, we introduce WavCaps, the first large-scale weakly-labelled audio captioning dataset, comprising approximately 400k audio clips with paired captions.
Ranked #1 on
Zero-Shot Environment Sound Classification
on ESC-50
(using extra training data)
RepCodec: A Speech Representation Codec for Speech Tokenization
To improve the performance of these discrete speech tokens, we present RepCodec, a novel speech representation codec for semantic speech tokenization.
