{"omc_version":"0.1","id":"espnet-owsm-ctc-v3-1-1b","name":"Owsm CTC v3.1 1B","provider":"ESPnet","released":"2024-02-23","licence":"Creative Commons Attribution 4.0","url":"https://huggingface.co/espnet/owsm_ctc_v3.1_1B","description":"Owsm CTC v3.1 is a tiny downloadable speech-to-text model from ESPnet that turns audio into written words at exceptional speed. It carries a permissive attribution-only licence and performs well on clean recordings, though its accuracy falls sharply on messier audio.","modalities":["audio"],"context_window":null,"capabilities":{"transcription":1.5},"x_llmap_capability_sources":{"transcription":{"suite":"asr-wer","suite_name":"Open ASR WER","rank":63,"of":74,"score":7.25,"higher_is_better":false,"variant":null}},"relationships":[{"type":"same-family","id":"espnet-owsm-ctc-v4-1b"}],"card_meta":{"created":"2026-08-01T22:45:45.761Z","created_by":"LLMap pipeline","last_updated":"2026-08-03T20:17:30.970Z","source":"https://llmap.ai/models/espnet-owsm-ctc-v3-1-1b","omc_spec":"https://github.com/openmodelcard/spec"}}