Models
One EmoKitten listens; three KittenTTS models speak. Model IDs are case-sensitive, so send the exact ID shown below.
Speech to text
| Model ID | Input | Output |
|---|---|---|
emokittenasr-realtime | Speech audio | Clean text plus emotion-aware transcript metadata |
The same model ID is used for upload and realtime STT.
For migrations from OpenAI clients, whisper-1, gpt-4o-transcribe,
gpt-4o-mini-transcribe, and gpt-realtime-whisper are accepted compatibility
aliases for the same hosted EmoKitten model. New integrations should use
emokittenasr-realtime; the aliases are not separate models.
Text to speech
| Model ID | Relative size | Price |
|---|---|---|
kitten-tts-nano-0.8 | Smallest | Free |
kitten-tts-micro-0.8 | Middle | Free |
kitten-tts-mini-0.8 | Largest and default | Free |
All three models support the voices and output formats listed in Generate speech.
Not sure which kitten to pick? Start with Mini, then benchmark Nano and Micro with your own text if model size or generation latency matters to your application. Small is the point here.
List models
curl --fail-with-body -sS \
https://api.kittenml.com/v1/models \
-H "Authorization: Bearer $KITTENML_API_KEY"
Use the catalog above as the supported inference contract. The model-list endpoint currently requires a key with ASR access and paid-ASR eligibility.