CoolFace
18 shown

datasets

Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.

modalityAllTextImageAudioTabularVideo
taskAllText generationQuestion answeringImage classificationSpeech
sizeAllUnder 1K1K – 10K100K – 1M1M – 10M
formatAllParquetJSONCSVWebDataset
sortTrendingMost likedMost usedRecently updatedNewest
Clear all
01RidheshBhati /Complete_Data_Source_100K_HOURS Multi-Language Audio Collection (100K Hours) This repository is physically reorganized for Absolute 100% Data Visibility. 🏗️ Global Consolidator Select your language subset to listen to high-quality waveform audio. All shards from legacy and modern pipelines are automatically routed here. audio1M<n<10M4 likes16k downloads5mo agoHugging Face02DynamicSuperb /SourceDetection_mb23-music_caps_4sec_wave_type_continuous Dataset Card for "SourceDetection_mb23-music_caps_4sec_wave_type_continuous" More Information needed audio1K<n<10K0 likes41 downloads3y agoHugging Face03ittailup /filtered_common_voice-hi-sourceaudio10K<n<100K0 likes34 downloads2y agoHugging Face0434data /nepali-music-source-seperation nepali-music-source-seperation Mirror of the exact bmcore v24 local holdout subset: 285 audio files. Benchmark label: real. This preserves the source benchmark label; it is not an independent label review. Source reference: https://huggingface.co/datasets/34data/v13-nepali-music-source-seperation. No new license or ownership claim is asserted by this mirror. Original source rights and restrictions remain applicable. Source revision reviewed:… See the full description on the dataset page: https://huggingface.co/datasets/34data/nepali-music-source-seperation.audion<1K0 likes32 downloads3d agoHugging Face05DynamicSuperb /SourceSeparation_libri2Mix_testaudio1K<n<10K1 likes29 downloads3y agoHugging Face06DynamicSuperb /SourceDetection_mb23-music_caps_4sec_wave_typeaudio1K<n<10K0 likes23 downloads3y agoHugging Face07Cybrpgs /sbpn-source-vote-disagreements-htdemucs-vs-ft-20260903gated Source-vote disagreements: htdemucs vs htdemucs_ft The SBPN chunker publishes, for every chunk, whichever of the original audio or the Demucs vocals scores higher on a librosa peak-to-10th-percentile RMS ratio, at a 0 dB threshold. Swapping the separation model therefore changes which audio ships, not merely its quality. Replaying that decision over 400 real chunks with htdemucs (shipped) and htdemucs_ft (vocals sub-model) produced 37 disagreements — a 9.25% flip rate. This… See the full description on the dataset page: https://huggingface.co/datasets/Cybrpgs/sbpn-source-vote-disagreements-htdemucs-vs-ft-20260903.audioaudio-classificationn<1K0 likes22 downloads19d agoHugging Face08Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-02gated gdrive-transcript1-audio-chunks-20260805-source-02 This gated dataset contains 1,095 MP3 speech and audio-event chunks from 50 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-02.audioautomatic-speech-recognition1K<n<10K0 likes13 downloads2mo agoHugging Face09anonymous-submission000 /bird3m-raw-sourcesaudion<1K0 likes11 downloads1y agoHugging Face10Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-04gated gdrive-transcript1-audio-chunks-20260805-source-04 This gated dataset contains 595 MP3 speech and audio-event chunks from 50 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-04.audioautomatic-speech-recognitionn<1K0 likes11 downloads2mo agoHugging Face11Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-01gated gdrive-transcript1-audio-chunks-20260805-source-01 This gated dataset contains 1,033 MP3 speech and audio-event chunks from 45 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-01.audioautomatic-speech-recognition1K<n<10K0 likes10 downloads2mo agoHugging Face12Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-06gated gdrive-transcript1-audio-chunks-20260805-source-06 This gated dataset contains 2,117 MP3 speech and audio-event chunks from 50 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-06.audioautomatic-speech-recognition1K<n<10K0 likes10 downloads2mo agoHugging Face13Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-03gated gdrive-transcript1-audio-chunks-20260805-source-03 This gated dataset contains 645 MP3 speech and audio-event chunks from 50 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-03.audioautomatic-speech-recognitionn<1K0 likes9 downloads2mo agoHugging Face1434data /v13-nepali-music-source-seperationaudion<1K0 likes8 downloads5mo agoHugging Face15Kppwdfgu1 /gdrive-transcript1-audio-chunks-20260805-source-05gated gdrive-transcript1-audio-chunks-20260805-source-05 This gated dataset contains 1,535 MP3 speech and audio-event chunks from 50 recordings. Access requires manual approval by the repository owner. Audio selection Every complete source recording was separated once with HTDemucs. Full vocal stems were stored remotely as 48 kHz mono 96k MP3; no per-chunk source separation was performed. Librosa SNR was calculated independently on aligned original and saved-vocals… See the full description on the dataset page: https://huggingface.co/datasets/Kppwdfgu1/gdrive-transcript1-audio-chunks-20260805-source-05.audioautomatic-speech-recognition1K<n<10K0 likes7 downloads2mo agoHugging Face16chen-829 /SourceSeparation_libri2Mix_testaudio1K<n<10K0 likes5 downloads3mo agoHugging Face17Batuka0901 /omni_source Omni source dataset Mongolian audio/text pairs used as source material for the omni training set. Dataset Statistics Total samples: 69 Total duration: 0h 15m 47s (0.26 h) Per-split breakdown Split Samples Total Duration Avg Duration train 69 0h 15m 47s (0.26 h) 13.73 s Single split (no held-out test set) -- this is source/reference data, not a benchmark split. audioautomatic-speech-recognitionn<1K0 likes3 downloads4mo agoHugging Face18Crixero /audio_sourceaudion<1K0 likes1 downloads1y agoHugging Face

Listings come live from the Hugging Face Hub API. CoolFace does not host these files.