OpenAI compatibility

KittenML supports the common OpenAI SDK workflows for upload transcription and streaming speech generation. Realtime STT uses OpenAI-style events on a KittenML-specific path.

Set this base URL for OpenAI HTTP SDKs:

https://api.kittenml.com/v1

Do not set the SDK base URL to a complete endpoint such as /v1/audio/transcriptions; the SDK appends that path itself.

Compatibility matrix

Capability	Status	Notes
POST /v1/audio/transcriptions	Supported	Works with OpenAI Python and JavaScript clients
model, file, language, response_format, stream	Supported	Use the documented values
json, verbose_json, text, srt, vtt	Supported	Includes KittenML metadata in JSON formats
Common OpenAI ASR model IDs	Compatibility aliases	They select the same hosted EmoKitten model; prefer emokittenasr-realtime
Upload SSE	Supported	Finite delta/done stream ending in [DONE]
POST /v1/audio/speech	Supported	Chunked binary audio by default; SSE is optional
model, input, voice, response_format, speed, stream_format	Supported	Use KittenML model IDs and voices
Realtime event names and common fields	Compatible style	Includes KittenML extension fields
Official OpenAI Realtime client connection	Not drop-in	The SDK hard-codes /v1/realtime; KittenML uses /v1/stt/realtime
WebRTC realtime transcription	Supported	Uses short-lived client tokens, SDP signaling, microphone media, and an oai-events data channel
Streaming TTS audio	Supported	stream_format: audio or stream_format: sse
Word-level timestamps, diarization, confidence scores	Not supported	Segment timestamps are available in verbose STT output

Compatibility covers the workflows above, not every field, model, endpoint, or behavior available from another provider.

For upload STT, the fields in the matrix are the accepted public contract. OpenAI options such as prompt, temperature, and word-level timestamp granularities do not affect KittenML inference and should be omitted. An unsupported response_format, empty file, or oversized file is rejected.

For TTS, send only the documented fields. Unsupported model IDs are rejected, and invalid input, voice, format, or speed values return field validation errors.

KittenML extensions

JSON and realtime transcript results can add:

• enriched_text
• accent
• tags
• clean_text
• segments
• request_id

OpenAI SDK response objects preserve these additional JSON fields.

Realtime clients

Use a standard WebSocket library with:

wss://api.kittenml.com/v1/stt/realtime?model=emokittenasr-realtime

See the tested Python and Node.js examples.

Browser applications can instead use the documented WebRTC transcription flow. KittenML follows the familiar short-lived-token and SDP exchange pattern, but uses the KittenML API hostname and documented signaling paths.