Documentation

Audio

Text to speech and audio transcription.

Speech

POST /audio/speech takes model and input (up to 4,096 characters). No other parameters are accepted. The response body is audio bytes, with the provider's content type. Billed per 1,000 characters, rounded up.

curl https://api.billionairestechnologies.com/api/v1/audio/speech \
  -H "Authorization: Bearer $FRIDAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"bt-voice","input":"Your order is ready."}' \
  --output speech.mp3

Transcription

POST /audio/transcriptions takes multipart form data with a file (audio, up to 25 MB) and model. The response is {"text": "..."}.

curl https://api.billionairestechnologies.com/api/v1/audio/transcriptions \
  -H "Authorization: Bearer $FRIDAY_API_KEY" \
  -F model=bt-ear \
  -F [email protected]

Transcription is billed per minute, and the duration is estimated from file size (assuming 128 kbps, minimum one minute). Such requests report x-friday-usage-basis: estimated.