Transcribe audio (for chat)

POST

/calls/transcribe

Converts base64-encoded audio into text. Used for voice input in the chat UI so users can speak instead of typing. Accepts common formats (webm, mp3, wav, m4a, ogg). Audio must be a base64 data URL. Not intended for call transcription — call transcripts are generated automatically during the call lifecycle.

Request

Authorization

JWT Bearer

Add the parameter

Authorization

to Headers

Example:

Authorization: ********************

Body Params application/jsonRequired

Examples

Responses

🟢200

application/json

Bodyapplication/json

Request Example

Shell

JavaScript

Java

Swift

curl --location '/calls/transcribe' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '{
    "audio": "string",
    "format": "wav"
}'

Response Example

{
    "transcription": "string"
}

Modified at 2026-07-23 18:02:16

Hide/delete an appointment from the list

Schedule a new call