Languages
English
English
The 0.x model cards do not list languages. Boubou speaks English with every checkpoint.
Source: KittenML/kitten-tts-mini-0.8
kitten-tts
Pipeline English 8 voices Mac · Linux with NVIDIAKittenTTS from KittenML: ONNX checkpoints from 15M to 80M parameters that run on the CPU. Each checkpoint has eight voices.
npx boubou@beta run kitten-tts:mini-0.8
Talk on this Mac. It needs Node.js 18 or later; Get started walks you through.
Talk on this Linux PC. It needs an NVIDIA GPU, and the first run also downloads a 3.62 GB engine that every pipeline voice shares; Get started walks you through.
npx boubou@beta serve kitten-tts:mini-0.8
Talk from a phone on the same Wi-Fi, or connect a robot or app. The model still runs on this Mac. Phones and robots walks you through.
Talk from a phone on the same Wi-Fi, or connect a robot or app. The model still runs on this PC. Phones and robots walks you through.
This version runs on Apple Silicon Macs only. On Linux, choose kitten-tts:mini-0.8.
- Download
- 5.66 GB5.34 GB
- Memory, estimated
- —
- GPU memory, measured
- 6–8.5 GiB
- Speaks
- English
- Voice
- Bruno
Recommended device
Apple Silicon Mac
- No memory estimate for this version yet
- macOS 15 or later
- Qwen3.5-4B, the language model Boubou maintains
- The voice runs on the CPU, with ONNX
Linux PC with an NVIDIA GPU
- NVIDIA driver 580.65.06 or later
- Tested on GPUs with 16 GB and 32 GB of memory
- x86-64, with glibc 2.39 or later, such as Ubuntu 24.04
- Qwen3.5-4B, the language model Boubou maintains, runs on llama.cpp
Memory needs are estimates. Boubou checks the machine before it downloads anything, and smaller machines still need testing.
Voices and languages
Everything the model can speak, as its own card or files list it. Black marks what Boubou uses.
- Boubou uses
- Available in the model
Voices
8 voices
Press a voice to hear it. Each clip is the pinned model reading one sentence in that voice’s language.
Voices on 0.8 checkpoints
Voice IDs on 0.1 and 0.2 checkpoints
expr-voice-2-mexpr-voice-2-fexpr-voice-3-mUsed in other versionsexpr-voice-3-fexpr-voice-4-mexpr-voice-4-fexpr-voice-5-mexpr-voice-5-f
On 0.8 checkpoints Bruno is the name of expr-voice-3-m, the voice Boubou uses on every checkpoint.
Source: KittenML/kitten-tts-mini-0.8
Versions 7
A version is one ready-to-run setup with its own command. Versions differ in size, language or the models they combine. Suggested ones also appear in the boubou menu.
| Version | What’s different | Speaks | Voice | Download | Mac | Linux |
|---|---|---|---|---|---|---|
| Largest: 80M parameters | English | Bruno | 5.66 GB | — | NVIDIA GPU | |
| 40M parameters | English | Bruno | 5.62 GB | — | NVIDIA GPU | |
| 15M parameters, full precision | English | Bruno | 5.64 GB | — | NVIDIA GPU | |
| 15M parameters, 8-bit weights | English | Bruno | 5.61 GB | — | NVIDIA GPU | |
| Earlier 0.1 release | English | expr-voice-3-m | 5.75 GB | — | NVIDIA GPU | |
| Earlier 0.1 release | English | expr-voice-3-m | 5.60 GB | — | NVIDIA GPU | |
| Earlier 0.2 release, 15M parameters | English | expr-voice-3-m | 5.60 GB | — | NVIDIA GPU |
How it works
You speak
VAD
Silero + Smart Turn v3.2
--vad sileroSTT
Parakeet TDT 0.6B v3
speech to textLLM
Qwen3.5-4B · 4-bit
--llm_backend mlx-lmTTS
kitten-tts · Bruno
--tts kittenIt answers
Boubou owns the microphone and speakers. Apple voice processing on a Mac, or WebRTC AEC3 on Linux, cancels echo, so you can talk over a reply. Boubou drops the queued audio and cancels the generation.
Swap a component
Boubou maintains one Hugging Face speech-to-speech 1.1.0 pipeline and pins every model in it. Add an upstream option after the model to replace one part.
Change the language model to any MLX model on Hugging Face
npx boubou@beta run kitten-tts:mini-0.8 --model_name mlx-community/Llama-3.2-3B-Instruct-4bit
Change speech recognition to Whisper
npx boubou@beta run kitten-tts:mini-0.8 --stt mlx-audio-whisper
Options work with serve too. Boubou does not check the models you name: they download from Hugging Face into your own Hugging Face cache. API modules send audio or text to the server you configure. These two examples are for a Mac: on Linux the pipeline stays offline, so options must name local paths.
- Used by Boubou
- Also runs in Boubou
VADVoice activity
--vadHears when you start and stop talking. Smart Turn v3.2 then decides whether a pause ends your turn.
silero
Not in Boubou’s runtime yet: firered
STTSpeech to text
--sttWrites down what you said. Parakeet is the default; Whisper versions detect the language.
parakeet-tdtmlx-audio-whisperwhisperqwen3-asrvllm-realtimeAPIopenaiAPIopenai-realtimeAPI
Not in Boubou’s runtime yet: faster-whisper, whisper-mlx, parakeet-unified, nemotron-streaming, orukeet, paraformer, sense-voice
LLMLanguage model
--llm_backendWrites the reply. Boubou runs pinned 4-bit Qwen3.5 models with mlx-lm; --model_name picks another one.
mlx-lmtransformerschat-completionsAPIresponses-apiAPI
TTSText to speech
--ttsSpeaks the reply while it is still being written. Every pipeline voice in the library is one of these.
qwen3kokoropocketphononkittensupertonicchatTTSfacebookMMSomnivoicedrama3openaiAPIminimaxAPI
License
Apache 2.0. Read the license. Weights keep their own licenses; check them before commercial use.