Languages
10 languages
ChineseEnglishJapaneseKoreanGermanFrenchRussianPortugueseSpanishItalian
English versions hear with Parakeet and answer in English as Aiden. Whisper versions detect your language and answer in it as Vivian.
qwen3-tts
Pipeline 10 languages 9 voices Mac only RecommendedQwen3-TTS CustomVoice as a full voice pipeline. Silero hears when you stop, a recognizer writes it down, Qwen3.5 answers and Qwen3-TTS speaks, all on your Mac.
npx boubou@beta run qwen3-tts:0.6b-customvoice
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 qwen3-tts:0.6b-customvoice
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.
qwen3-tts runs on Apple Silicon Macs only. Other models run on Linux.
- Download
- 7.41 GB
- Memory, estimated
- 12 GiB
- Speaks
- English
- Voice
- Aiden
Recommended device
Apple Silicon Mac
- 16 GB of memory or more
- macOS 15 or later
- Qwen3.5-4B, the language model Boubou maintains
- The voice runs on MLX on Apple Silicon
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
9 voices
Press a voice to hear it. Each clip is the pinned model reading one sentence in that voice’s language.
CustomVoice speakers
Each speaker is labelled with its native language, where the card recommends using it. CustomVoice checkpoints offer only these speakers.
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 |
|---|---|---|---|---|---|
| The default, with the smaller voice model | English | Aiden | 7.41 GB | 16 GB+ | |
| Larger voice model | English | Aiden | 8.28 GB | 32 GB+ | |
| Hears and answers in your language | Your language | Vivian | 6.52 GB | — | |
| Larger voice, answers in your language | Your language | Vivian | 7.38 GB | — | |
| Larger language model: Qwen3.5-9B | English | Aiden | 11.2 GB | 64 GB+ | |
| Largest language model: Qwen3.5-27B | English | Aiden | 21.3 GB | 128 GB+ | |
| Lightest: Whisper tiny and Qwen3.5-2B | English | Aiden | 3.75 GB | 16 GB+ |
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
qwen3-tts · Aiden
--tts qwen3It answers
Boubou owns the microphone and speakers. Apple voice processing 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 qwen3-tts:0.6b-customvoice --model_name mlx-community/Llama-3.2-3B-Instruct-4bit
Change speech recognition to Whisper
npx boubou@beta run qwen3-tts:0.6b-customvoice --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.
- 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.