AsyncAPI channel · Krisp
· Krisp Voice Translation WebSocket API
/vt
A single bidirectional translation session. The client sends exactly one JSON config message, then streams binary audio; the server streams synthesized translated audio back plus JSON transcript, translation, and error events.
Once the WebSocket connection is established, the client MUST send a single JSON configuration message as the first message, before sending any audio.
send
sendAudio
Raw binary audio frames in the encoding declared in config.audio.format. Pace in real time — one chunk per input_frame_duration ms (640 bytes per 20 ms at 16 kHz mono s16le).
receive
receiveEvents
Synthesized translated speech plus transcript, translation, and error events.
Messages
sessionConfig
Initial client configuration
Content-Type: application/json
inboundAudio
Raw PCM (pcm_s16le) or Opus audio, 16 kHz mono, as a binary WebSocket frame.
Content-Type: application/octet-stream
translatedAudio
Translated speech in the same encoding as the inbound stream, 16 kHz mono, binary frame.
Content-Type: application/octet-stream
transcript
Source transcript event
Content-Type: application/json
translate
Translated transcript event
Content-Type: application/json
error
Emitted when the server encounters an issue processing the request.
Content-Type: application/json
About AsyncAPI
The AsyncAPI specification describes event-driven APIs the way OpenAPI describes request/response APIs. A channel is the named pipe — a webhook URL, a Kafka topic, a WebSocket route, an MQTT subject — that producers and consumers publish or subscribe to. Each channel carries one or more messages with structured payloads, and an operation declares whether a given party sends or receives on that channel.