{"omc_version":"0.1","id":"soundsgoodai-zipformer-transducer-xl-290m","name":"Zipformer Transducer XL 290M","provider":"Sounds Good AI","released":"2026-05-12","licence":"Creative Commons Attribution-NonCommercial 4.0","url":"https://huggingface.co/soundsgoodai/Zipformer-transducer-XL-290M","description":"Zipformer Transducer XL is a tiny downloadable speech-to-text model from Sounds Good AI that turns English audio into written words. It is extremely fast on benchmark hardware and accurate on clean recordings, but its error rate rises sharply on meetings, podcasts and accented speech.","modalities":["audio"],"context_window":null,"capabilities":{"transcription":2.5},"x_llmap_capability_sources":{"transcription":{"suite":"asr-wer","suite_name":"Open ASR WER","rank":45,"of":74,"score":6.175714286,"higher_is_better":false,"variant":null}},"relationships":[],"card_meta":{"created":"2026-08-01T22:45:55.178Z","created_by":"LLMap pipeline","last_updated":"2026-08-03T20:17:39.664Z","source":"https://llmap.ai/models/soundsgoodai-zipformer-transducer-xl-290m","omc_spec":"https://github.com/openmodelcard/spec"}}