Skip to main content
Turn your audio files into clean, readable text using our optimized speech-to-text engine.

​
Endpoint

POST https://api.llm.kiwi/v1/audio/transcriptions

​
Request Parameters

ParameterTypeRequiredDescription
filefileYesThe audio file to transcribe (supported: mp3, wav, m4a, etc.).
modelstringYesUse default for our standard transcription engine.
languagestringNoISO-639-1 language code of the input (e.g., en, fr).
promptstringNoOptional text to guide the style or provide context.

​
Example Usage

curl https://api.llm.kiwi/v1/audio/transcriptions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: multipart/form-data" \
  -F file="@recording.mp3" \
  -F model="default"

​
Best Practices

  • Audio Quality: High-quality audio with minimal background noise yields significantly better results.
  • Supported Formats: We recommend .mp3 or .m4a for a good balance of quality and file size.
  • File Size Limits: Keep audio files under 25MB for optimal performance.