Cartesia STT WebSocket API
Realtime transcription over WebSocket at wss://api.cartesia.ai/stt/websocket. Clients stream raw binary audio in 100ms chunks and issue finalize/close text commands; the server returns incremental and final transcript messages with word-level timing. A companion "Auto" variant at /stt/turns/websocket adds built-in turn detection for conversational agents.
Cartesia STT WebSocket API is one of 21 APIs that Cartesia publishes on the APIs.io network, described by an AsyncAPI event-driven specification.
Tagged areas include WebSocket, Speech to Text, Realtime, and Turn Detection. The published artifact set on APIs.io includes API documentation, an API reference, and an AsyncAPI specification.
This event API exposes 4 operations across 2 channels, and defines 24 schemas. It is described by AsyncAPI 2.6.0, at version 2026-03-01.
The API connects through 2 servers: api.cartesia.ai/tts/websocket?cartesia_version=2026-03-01 (wss), api.cartesia.ai/stt/websocket?model=ink-whisper&encoding=pcm_s16le&sample_rate=16000&cartesia_version=2026-03-01 (wss).
Metadata
The identity and technical contract details declared by the specification.
Channels & Operations 4
Across 2 channels, the API surfaces 4 operations. Each is listed below with its action, channel, parameters, and response codes.
Messages & Schemas 24
The contract defines 24 schemas that model the data carried by its messages. The most detailed are TtsGenerationRequestPayload (13 properties), ErrorPayload (9 properties), TtsAudioChunkPayload (6 properties), FlushDonePayload (6 properties). Each schema is shown below with its type and property counts.
Specification
The full machine-readable AsyncAPI contract behind this narrative.
Source
More from Cartesia 12
Other APIs Cartesia publishes across the network.