Beta
Transcribe Arabic audio through one simple API
A direct transcription contract
Upload audio as multipart form data and receive a small JSON response with the transcript and billed audio duration.
Built into the same voice platform
Use one account, API key, prepaid balance, and usage view for Arabic speech generation and transcription.
OpenAI-compatible endpoint shape
Call POST /v1/audio/transcriptions with familiar authentication and multipart request conventions.
Arabic audio in, usable text out
The transcription endpoint is designed for application workflows rather than a manual upload portal. It accepts common audio files, returns JSON, and bills by actual audio duration at a published rate.
- 01
Prepare the recording
Use a WAV, MP3, FLAC, OGG, or M4A file with clear Arabic speech.
- 02
Upload with your API key
Send the file to POST /v1/audio/transcriptions as multipart form data.
- 03
Read the transcript
Use the returned text in search, media workflows, support tools, or downstream language processing.
Questions developers ask
Which audio formats can I transcribe?
The platform accepts WAV, MP3, FLAC, OGG, and M4A uploads. The public API reference documents the current file and response contract.
Does the response include timestamps?
No. The current response has no word or segment timestamps, including when verbose_json is requested. Use json for the transcript and measured usage.
How is Arabic speech to text priced?
Transcription costs $0.50 per hour of audio and is deducted from the same prepaid USD balance used by text to speech.