Speech Transcription with parakeet-tdt-0.6b-v3 🦜

This API-first demo uses parakeet-tdt-0.6b-v3 to transcribe uploaded audio files with word-level timestamps.

Upload an audio file to get word-level timestamps in a structured JSON response. This Space is optimized for file uploads and API usage only; microphone/live transcription, transcript downloads, and preview playback have been removed.

🎙️ Model Card | 🧑‍💻 NeMo Repository

Example Audio Files (Click to Load)

Structured transcription response