Whisper — распознавание речи в текст

★ 70k+

Whisper is OpenAI's open speech-recognition model that transcribes audio to text with high accuracy in many languages and can also translate speech. Reach for it when you need to transcribe audio or video to text locally: transcribe interviews, lectures, podcasts, and meeting recordings, get subtitles, and work with multilingual audio — on your own hardware, from open weights. It is the reference ASR model on which many accelerated and applied tools are built. Its focus is a base open speech-recognition model (speech → text) for running and embedding, not a finished app with a UI (buzz), a fast production engine (faster-whisper), or transcription with diarization (whisperX): the value is accurate multilingual transcription as a foundation. Larger models are more accurate but more resource-hungry. It suits developers and researchers who need a reference ASR model for their tasks.