100,000-hour multilingual speech corpus
A collection of audio recordings and texts in 6 languages, including English, Hindi, Bengali, and Tamil.
- Assamese
- Bangla
- English
- +9
Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.
Answered by Gemini 4 · the chat is kept only in this tab
A collection of audio recordings and texts in 6 languages, including English, Hindi, Bengali, and Tamil.
Female ASMR voice recordings in AAC format with Whisper word-level alignment and non-speech sound tags for ASR and generation tasks.
Over 133,000 hours of voiced books across 86 languages for training speech recognition and synthesis systems.
Audio dataset of news recordings with text transcripts for rare dialects.
A large-scale audio dataset and transcripts of Legislative Council meetings in Cantonese, featuring raw and segmented recordings.
A public gated audio-text dataset of over 150k samples (~158 hours) for speech tasks.
Classic dataset of audio recordings with various speech and vocal emotions for emotion recognition.
Millions of audio recordings from coral reef restoration sites with sound type annotations for bioacoustics.
A classic 10-class urban sound audio dataset with pre-split cross-validation folds.
A small dataset of bird song recordings from Doñana National Park for species detection and classification.
Over 362k instrumental tracks from Jamendo, enhanced with generated text descriptions.
Nearly 12,000 hours of Russian speech with consensus transcriptions from seven ASR models and added punctuation.
Validation and test splits of the CoVoST2 speech translation dataset with embedded audio files.
16.1 hours of conversational speech with manual word-level time stamps for forced alignment.
156 hours of Urdu and Pakistani English speech audio recordings with 12 metadata parameters for ASR.
A copy of the Daimon-Infinity audio repository from ModelScope with unchanged structure and files for audio tasks.
Over one million hours of English audio from Archive.org for ASR and audio classification.
Ready-to-use head, antenna, and body animations with audio tracks for Reachy Mini emotional expressions.
Augmented audio recordings of short voice commands in 15 languages for keyword spotting systems.
Audio dataset with Vietnamese transcriptions (32,267 fragments, 16 kHz) for speech recognition.
Combined corpus of 16 datasets with video, audio, and text for recognizing emotions, humor, and human state.
A re-upload of the well-known first-person video benchmark with audio for quick downloading.
654k wav files of Genshin Impact character voice lines in Chinese, English, Japanese, and Korean.
About 585 hours of clean multi-speaker English speech (24 kHz) with texts for speech synthesis.