NOTSOFAR-1 single-device meeting recordings
Audio data and baseline solutions from Microsoft for remote meeting speech transcription.
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
Audio data and baseline solutions from Microsoft for remote meeting speech transcription.
Audio recordings of everyday sentences in Irish English.
Over 100k first-person CS2 rounds with 720p video, stereo audio, and tick-synchronized data for all 10 players.
Dataset with text, images, audio, and tables for multimodal models (100K-1M entries).
A small sample of two-way conversations with separated speaker tracks for ASR.
Unified multimodal dataset combining images, audio, and text from public sources.
Audio recordings of normal and anomalous machine operation from MIMII and ToyADMOS for training acoustic fault detectors.
Audio and text for ASR and TTS across dozens of African languages—a large open speech corpus from Google.
An audio-video benchmark with masks for evaluating referring audiovisual segmentation quality.
49K English and Chinese questions about cats and dogs with images, audio, video, and text.
Synchronized video, audio, photo, and sensor data for human action recognition.
Around 893,000 chapters from 96,000 Russian audiobooks with audio files and descriptions.
ASR tests with audio and text, sorted by length and saved in parquet format.
Read Khmer speech based on Tipitaka texts for ASR and TTS tasks with manual access moderation.
From 10K to 100K audio dialogues in English and Chinese for training conversational models with multiple speakers.
450+ non-trivial questions with single-answer targets for testing tool-using and search-equipped LLMs.
Speech corpus divided into clean and noisy speech for training and evaluating ASR systems.
Large-scale extraction of texts, images, audio, and video from open datasets in a unified parquet format.
A physically verified multimodal benchmark for bird sound attribution.
Audio recordings of individual Cantonese syllables generated by Amazon, Apple, Google, and Microsoft engines.
Hundreds of thousands of audio clips from 14K hours of speech with timestamp annotations and summaries.
271 audio files featuring acoustic impulse responses for speech augmentation and reverberation simulation.
16 kHz Russian speech audio recordings with manual annotations from the crowdsourced part of the Golos corpus for ASR tasks.
A large heterogeneous English speech corpus with transcriptions and ready-made subsets for training ASR models.