Voicebox

Open source alternative to ElevenLabs, Google Cloud Text-to-Speech and Murf AI

58.1K stars
MIT
Active this month
Last commit 15 days ago

Last verified Oct 21, 2026

Voicebox is an open-source, local-first voice synthesis studio for cloning voices, generating speech, and building voice-powered apps.

Visit website GitHub58.1K

Read the full breakdownWho it is for, how it works, trade-offs and the FAQ

who it's for

Who Voicebox is for#

Creators replacing hosted TTS tools

Use Voicebox when creators want local speech generation and voice profile control instead of browser-only subscriptions.

Skip if:

Skip it if the team needs managed commercial voice licensing, studio support, and guaranteed production SLAs.

Developers adding voice to agents

Use Voicebox when local agents, prototypes, or desktop workflows need spoken output through an inspectable tool.

Skip if:

Skip it if the app needs a scalable hosted voice API managed by a vendor.

the problem

The problem it solves#

Hosted voice tools make speech generation easy, but they often require uploading scripts, samples, and voice data to a third-party service. That is uncomfortable for creators, developers, and teams working with unreleased products, private narration, or sensitive meeting notes. Voice cloning also raises consent and rights questions that a tool cannot solve for the user.

Voicebox targets people who want voice generation and dictation on their own machine. The core value is local control over voice profiles, TTS engines, speech generation, dictation, and agent voice output rather than a browser-only voice subscription.

how Voicebox solves it

How it solves it#

Local desktop voice studio

Run voice cloning, text-to-speech, dictation, and agent voice output from a desktop app instead of sending every workflow through a hosted voice dashboard.

Multiple TTS engines

Voicebox presents several text-to-speech engines behind one workflow. Developers and creators can compare voice quality and latency without rebuilding the studio around each engine.

Dictation into other apps

Use speech input beyond the Voicebox window by dictating into existing applications. That makes the tool relevant for daily writing and agent workflows, not only audio export.

Agent voice output and API paths

Voicebox exposes ways for agent tools to speak through a cloned or selected voice. That fits local assistant demos, accessibility experiments, and voice-enabled developer workflows.

strengths · trade-offs

Strengths and trade-offs#

Strengths

Local-first alternative to hosted voice SaaSUnlike ElevenLabs-style hosted workflows, Voicebox is designed to run on the user’s machine. That is useful when voice data, prompts, or generated speech should stay local.
Combines creation and daily inputVoicebox is not only a voice-cloning demo. It connects cloning, TTS, dictation, and agent speech, which makes it more useful for repeated desktop workflows.

Trade-offs

-Voice rights and model quality remain user responsibilitiesLocal software does not remove consent, likeness, copyright, or quality-review obligations. Teams should define voice-use rules before cloning real people or publishing generated speech.

versus alternatives

Voicebox vs alternatives#

Voicebox vs ElevenLabs

Voicebox is the better fit when a creator or developer wants local voice cloning, TTS, dictation, and agent speech without routing every workflow through a hosted account. ElevenLabs is stronger when a team needs managed voice infrastructure, commercial licensing support, collaboration, and scalable hosted APIs. Choose Voicebox for local control; choose ElevenLabs for managed production voice services.

tech stack · detected from GitHub

What it's built on#

Languages: PythonRustTypeScript
Frameworks: Next.jsReact

frequently asked

FAQ#

What does Voicebox replace?

Voicebox can replace parts of ElevenLabs, WisprFlow, and hosted dictation tools when the need is local voice cloning, speech generation, dictation, or agent voice output.

Is Voicebox self-hosted?

Voicebox is primarily a local desktop app, not a server product. The official site describes macOS, Windows, and Linux downloads that run on the user’s machine.

What license does Voicebox use?

The OSA item record lists MIT. Review the upstream repository license and any model-specific terms before commercial use or redistribution.

also worth a look

Similar open-source tools#

VoiceStudio

Alternative toElevenLabs

Voice cloning, dubbing, and transcription that runs fully offline

78,615 GitHub stars
Licence AGPL-3.0

supertonic

Alternative toElevenLabs

Instantly convert text to speech with customizable options

13,772 GitHub stars
Licence MIT

VoxCPM

Alternative toElevenLabs

Tokenizer-free multilingual text-to-speech with voice cloning

39,615 GitHub stars
Licence Apache-2.0

Hume AI

Alternative toElevenLabs

Empathic voice AI SDK with open source emotional models

84 GitHub stars
Licence MIT

Openfilm

Alternative toRemotion

AI video agent for coding agents; films are local web folders.

1,850 GitHub stars
Licence MIT

SOMA

Alternative toZoo.dev

Parametric 3D modeling via CLI agent: describe, generate, iterate.

162 GitHub stars
Licence GPL-3.0

by category

Open source tools in these categories#

404

Featured

Omnara

Alternative toClaude Managed Agents

Open-source agent deployment API. Self-host or use Omnara Cloud.

2,957 GitHub stars
Licence Apache-2.0
#AI Agents

Inbox Zero

Alternative toSuperhuman

AI email assistant that drafts replies and clears your inbox

12,712 GitHub stars
Licence AGPL-3.0
#Email

rakazo

Alternative toGrok Bot

AI teammates you own: your keys, your model, your machine.

4,904 GitHub stars
Licence Apache-2.0
#AI Agents

Feature your project

The projects we add and every new guide. Free, unsubscribe any time.

Find open source alternatives to paid software

Who Voicebox is for#