What you can do in Transcribe

Speech to text

• Upload MP3 or WAV and get a transcript.
• Optional model choice on the API: auto, whisper, or ai4bharat.
• Background jobs (async_mode) so you can poll history instead of waiting.
• Live Recording: browser microphone, 16 kHz mono PCM over WebSocket.

Languages

• Auto language handling for English and Indian languages.
• Languages tab lists supported codes and names (Indian languages marked for the Indic model).

History & export

• Transcript History: UUID, filename, size, duration, model, language, status, text.
• Filters: All, Completed, Processing, Queued, Failed.
• Export CSV / PDF.

Project & account

• Studio project selector (example: AI Project).
• Dark/light theme.
• You only see your own transcription jobs in the Transcribe APIs.

Reporting & control

• Dashboard: files processed, jobs, minutes, storage.
• Studio Administration: Projects, Users, Audit Logs (platform-wide).

Optional: API

Same capabilities exist as REST and WebSocket (Swagger). See API if you are not using the Studio UI.

Next: How it works | Back: Getting Started