sts
Real-time speech-to-speech translation channel.
Handshake
WSS
wss://dummy.api.kotobatech.ai/sts
Authentication
AuthorizationBasic
HTTP Basic authentication. Set device_id as the username and api_key as the password. Send it via the Authorization header during the WebSocket handshake.
Send
clientSendSessionUpdate
Event that must be sent after the WebSocket connection is established
and before sending audio. Sets the input audio format, sampling rate,
source language, and target language.
OR
clientSendAudio
Sends a Base64-encoded audio chunk (as a JSON frame). Alternatively,
PCM/Opus data can be sent directly as binary frames.
OR
clientCommitAudio
Sent when ending audio transmission (e.g., when the microphone is
turned off). The server processes all remaining audio in the buffer.
Receive
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.
OR
serverReceiveEvents
Receives session creation, text deltas, audio deltas, commit
completion notifications, errors, and so on.