NTU ML2021 ASR and Speech Translation
Recordings from a National Taiwan University machine learning course pair Mandarin and English lecture speech with text. The corpus includes spoken switches into English technical terms.
Visit official source (opens in a new tab)Research use
Evaluate recognition of Chinese and English within the same technical lecture. Use English translations for speech translation tasks.
Limitations
The release focuses on one course and its technical vocabulary. Estimate speaker and topic transfer on separate data.
License and access
- Dataset license
- MIT (publisher data card) (opens in a new tab)
- Access status
- Direct access
Select the train, development or test split. Use the transcription field as the ASR reference.