Speech-to-text (Whisper v3) · x402.telnyx.com
Speech-to-text (OpenAI-compatible /v1/audio/transcriptions…
Multipart). Whisper large-v3-turbo, 100 languages, verbose_json with segment timestamps + duration. wav/mp3 up to 5 min / 12 MB. $0.003 USDC per minute (Telnyx list rate, zero margin), metered per second via the upto scheme (min $0.002 network floor ≈ 40s), or flat $0.015 exact.
Sample input
input4 keys
body3 keys
file"<multipart part "file": wav or mp3 bytes>"
model"whisper-1"
response_format"verbose_json"
bodyType"form-data"
method"POST"
type"http"
inputobject, 4 keysSample response
response3 keys
duration4.05
segments1 item
0{ end, id, start, … }4 keys
text"Hello from the x402 gateway."
responseobject, 3 keysTrack record
- Paid calls, last 30 days
- 12
- Buyers, last 30 days
- 1
- Last paid call
Reported by the x402 directory.