Open-source, local-first voice studio for voice cloning, text-to-speech, dictation and agent voice output.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Voicebox is a desktop app that combines voice cloning and speech generation with speech-to-text dictation, all running locally. It is positioned as a free, open-source alternative to ElevenLabs (output) and WisprFlow (input). The README states that models, voice data and captures stay on the user's machine.
It bundles eight TTS engines (Qwen3-TTS, Qwen CustomVoice, Qwen VoiceDesign, LuxTTS, Chatterbox Multilingual, Chatterbox Turbo, HumeAI TADA and Kokoro) covering 23 languages. Features include zero-shot cloning from a short reference sample, 50+ preset voices, paralinguistic tags such as [laugh] and [sigh], auto-chunking for long scripts, post-processing effects (pitch, reverb, delay, chorus, compression, filters), and a multi-track Stories timeline editor for podcasts and conversations.
For input, it offers a global dictation hotkey (push-to-talk or toggle) with Whisper-based transcription and auto-paste on macOS. A REST API and built-in MCP server let agents such as Claude Code, Cursor and Cline speak through a cloned voice via a voicebox_speak tool, and voice profiles can carry personas used with a bundled local LLM. It is built with Tauri (Rust) and runs on macOS (MLX/Metal), Windows (CUDA), Docker, AMD ROCm and Intel Arc. Linux pre-built binaries are not yet available, so Linux requires building from source.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Open-source, free local voice stack offering voice cloning, speech generation and dictation, pitched as ElevenLabs plus WhisperFlow. Includes a stories timeline editor, audio effects pipeline and transcription; the demo shows Qwen 1.7B generating speech.
Closest neighbours by meaning, not just by category.
Model · Resemble AI
Open-weights expressive text-to-speech model from Resemble AI with voice cloning, fine-tuned from LTX-2.3 audio.
API · ElevenLabs
AI audio platform and API for text-to-speech, speech-to-text, music, sound effects, voice cloning and voice agents.
Model · Edge0
Open-weights 0.17B-parameter multilingual text-to-speech model with zero-shot voice cloning, runs on CPU via ONNX.
Model · Rednote
Open-source 2B-parameter zero-shot voice-cloning text-to-speech model from Rednote's hilab team.
Model
Expressive text-to-speech model with zero-shot voice cloning, derived from the LTX 2.3 audio model.
Model
Compact 0.6B open text-to-speech model with zero-shot voice cloning and support for 23 languages.
