Skip to content
SPECIFICTION
Get listed

Voicebox

voicebox.sh

From $1/mo Free plan Open source WebWindowsmacOSLinuxiOSAndroid
Visit voicebox.sh
Spec score 9.4 #2 of 233 in Text-to-speech tools

What it is

Voicebox is a free, open-source local AI voice studio for voice cloning, speech generation, dictation, and agent voice output. Its desktop app can clone a voice from as little as three seconds of audio, supplied by file, microphone recording, or system capture. The maker lists seven text-to-speech engines for cloning, preset voices, and delivery control. A timeline editor supports multi-voice narratives, arranging tracks, trimming clips, and mixing conversations; effects include pitch shift, reverb, delay, and compression. Dictation uses a global shortcut to send transcripts to the focused text field or clipboard, with local Whisper models supporting 99 languages. MCP-aware agents such as Claude Code, Cursor, and Cline can speak through Voicebox, and a local REST API provides speech generation and related functions. The privacy policy says the offline desktop app requires no account and does not send recordings, profiles, transcripts, generations, or settings to the maker. Local is free with unlimited local generations and captures. Cloud and mobile sync are marked as coming soon, and cloud pricing and limits are not final.

Spec sheet

What it does
Voicebox is a local AI voice studio for voice cloning, text-to-speech, dictation, transcription, and agent speech.Read 14 Oct 2026 · voicebox.sh
Product
Voicebox is a free, open-source local AI voice studio for voice cloning, speech generation, dictation, and agent voice output.Read 2 Oct 2026 · voicebox.sh
Free plan
YesRead 20 Sep 2026 · voicebox.sh
API
Voicebox exposes a local REST API for speech generation, voice profiles, model status, generation history, and health checks.Read 2 Oct 2026 · voicebox.sh
Maker
The download page identifies Jamie as the maintainer and says Voicebox is a spare-time side project; the site footer attributes it to Spacedrive Technology Inc.Read 2 Oct 2026 · voicebox.sh
Agent integration
The built-in MCP server lets Claude Code, Cursor, Cline, Windsurf, and other MCP clients call Voicebox speech tools.Read 14 Oct 2026 · voicebox.sh
Agent integrations
MCP-aware agents including Claude Code, Cursor, and Cline can speak through Voicebox using a cloned voice; the maker also describes a POST /speak endpoint for other clients.Read 2 Oct 2026 · voicebox.sh
API access
YesRead 20 Sep 2026 · voicebox.sh
API endpoint
The documented local server runs at http://127.0.0.1:17493 and exposes OpenAPI documentation at /docs.Read 14 Oct 2026 · voicebox.sh
Architecture
Voicebox uses a Tauri desktop shell, React and TypeScript frontend, and Python FastAPI backend communicating over localhost HTTP.Read 14 Oct 2026 · docs.voicebox.sh
Audio effects
Its effects pipeline supports pitch shift, reverb, delay, compression, and saved presets.Read 14 Oct 2026 · voicebox.sh
Cloud security
The announced cloud backup encrypts library data on users’ devices, while servers store ciphertext and cannot decrypt it.Read 2 Oct 2026 · voicebox.sh
Cloud status
Voicebox Cloud is marked coming soon, and the pricing page says cloud pricing and limits are not final.Read 2 Oct 2026 · voicebox.sh
Commercial use
YesRead 20 Sep 2026 · voicebox.sh
Company facts: acquisition record
Crunchbase lists Voicebox Technologies Corporation as acquired by Nuance Communications and operating status closed.Read 14 Oct 2026 · crunchbase.com
Company facts: similarly named entity
LinkedIn lists a separate Voicebox Technologies Corporation in Bellevue, Washington, founded in 2001, with 201–500 employees.Read 14 Oct 2026 · linkedin.com
Company identity
The Voicebox website footer identifies the maker as Spacedrive Technology Inc.Read 14 Oct 2026 · voicebox.sh
Dictation
A global keyboard shortcut records speech and places the transcript into a focused application field or the clipboard.Read 14 Oct 2026 · voicebox.sh
Documentation
The documentation covers installation, Docker deployment, quick start, GPU acceleration, dictation, cloning, MCP, stories, troubleshooting, API reference, architecture, building, and contributing.Read 14 Oct 2026 · docs.voicebox.sh
Languages
23Read 20 Sep 2026 · voicebox.sh
Licence
The source repository is released under the MIT licence, permitting use, modification, distribution, sublicensing, and sale subject to the notice requirement.Read 14 Oct 2026 · github.com
Limit: cloning setup
Dave Swift reported that models must be installed before cloning, clipped recordings can make cloning fail, and the app uses the operating system's selected microphone.Read 14 Oct 2026 · daveswift.com
Limit: commercial polish
HysenLabs characterized Voicebox as unsuitable for users needing a fully polished commercial product or prebuilt Linux binaries.Read 14 Oct 2026 · hysenlabs.com
Limit: compliance
HysenLabs noted that the project claims no compliance certification despite local processing being suitable for privacy-sensitive workflows.Read 14 Oct 2026 · hysenlabs.com
Limit: effects bug
Dave Swift reported that built-in effects were unimpressive and saving customized presets produced an error.Read 14 Oct 2026 · daveswift.com
Limit: hardware dependence
Dave Swift reported that local hardware directly affects generation speed and that stronger machines produce faster results.Read 14 Oct 2026 · daveswift.com
Limit: remote API security
Limpid Reviews warned that the localhost REST API is unauthenticated by default and could be risky if exposed on shared or networked machines.Read 14 Oct 2026 · limpidreviews.fyi
Linux availability
The installation guide says Linux builds are coming soon and currently blocked by GitHub runner disk-space limits.Read 14 Oct 2026 · docs.voicebox.sh
Local privacy
The desktop app works offline without an account and does not transmit recordings, profiles, transcripts, generations, or settings to the maker, according to its privacy policy.Read 2 Oct 2026 · voicebox.sh
Mobile and Docker
The maker describes cloud sync across desktop and mobile, and its documentation lists Docker deployment; the cloud and mobile offering is marked coming soon.Read 2 Oct 2026 · voicebox.sh
Model downloads
Models are downloaded from Hugging Face on first use and cached locally; documented model sizes range from about 300 MB to 8 GB.Read 14 Oct 2026 · docs.voicebox.sh
Notable news: release fix
GitHub release notes dated April 22, 2026 describe version 0.4.5 fixing offline-mode model-load crashes and eliminating a network timeout pause.Read 14 Oct 2026 · github.com
Praise: local privacy
Independent reviewer Dave Swift praised that processing is local, so voice data does not leave the computer and there are no subscriptions or account requirements.Read 14 Oct 2026 · daveswift.com
Praise: voice quality
Dave Swift found Qwen3 audio clear and natural and considered Voicebox impressively capable for a free open-source application.Read 14 Oct 2026 · daveswift.com
Rating on Scam Detector
13.4 out of 100 from 53 algorithmic factorsRead 14 Oct 2026 · scam-detector.com
Repository activity
The GitHub repository page shows 388 issues and 118 pull requests.Read 14 Oct 2026 · github.com
REST API
A localhost REST API exposes generation, cancellation, profiles, model status, history, health, and speech endpoints without API keys or usage fees.Read 14 Oct 2026 · voicebox.sh
Security policy
The project accepts vulnerability reports at [email protected], says releases are cryptographically signed, and states local audio stays on the machine unless remote mode is enabled.Read 14 Oct 2026 · github.com
Security posture
Local mode binds to 127.0.0.1 without authentication; planned remote protections include API keys, accounts, rate limiting, and HTTPS.Read 14 Oct 2026 · docs.voicebox.sh
Self-hosting
The backend can run locally or on a separate Docker or bare-host machine for remote-mode connections.Read 14 Oct 2026 · docs.voicebox.sh
Speech engines
The maker lists seven text-to-speech engines, with models for cloning, preset voices, and delivery control.Read 2 Oct 2026 · voicebox.sh
Stories editor
The Stories editor arranges multi-voice narratives on a timeline, with track trimming and conversation mixing.Read 14 Oct 2026 · voicebox.sh
Supported desktop systems
Downloads are listed for macOS Apple Silicon and Intel, Windows 64-bit, and Linux build from source.Read 2 Oct 2026 · voicebox.sh
System requirements
Minimum requirements are macOS 11 or newer, Windows 10 or newer, 8 GB RAM, 5 GB free storage, and a modern multi-core CPU; 16 GB RAM and a CUDA GPU are recommended.Read 14 Oct 2026 · docs.voicebox.sh
Transcription
Whisper models provide local speech-to-text in 99 languages, with optional local LLM refinement.Read 14 Oct 2026 · voicebox.sh
TTS engines
The application includes Qwen3-TTS, Qwen CustomVoice, LuxTTS, Chatterbox, Chatterbox Turbo, TADA, and Kokoro engines.Read 14 Oct 2026 · docs.voicebox.sh
Use cases
The maker describes use for game dialogue, apps and agents, accessibility readouts, audiobooks, podcast intros, and scripts or tools.Read 2 Oct 2026 · voicebox.sh
Voice cloning
It can clone a voice from a few seconds of audio supplied by upload, microphone recording, or system-audio capture.Read 14 Oct 2026 · voicebox.sh

Plans and prices

Local Free free forever No account required · all TTS engines · voice cloning · dictation and capture · MCP/personality features · unlimited local generations Read 14 Oct 2026 · voicebox.sh
Cloud $12 a year launch price; coming soon End-to-end encrypted backup · 25 GB storage · up to 5 devices · 30-day version history · sync across desktop and mobile Read 14 Oct 2026 · voicebox.sh
Studio $48 a year placeholder; pricing TBD; coming soon 250 GB encrypted storage · unlimited devices · one-year version history · priority support Read 14 Oct 2026 · voicebox.sh

Every Voicebox plan and price →

Strong points and limits

Strong points

  • Voice cloning can start from three seconds of audio.
  • Seven text-to-speech engines are listed.
  • Local Whisper dictation supports 99 languages.
  • Offline use without an account.
  • Free plan includes unlimited local generations and captures.

Limits

  • Cloud backup and mobile sync are coming soon.
  • Cloud pricing and limits are not final.
  • No free trial is offered.
  • Linux downloads require building from source.

Who it suits

Voicebox may suit people creating game dialogue, app or agent speech, accessibility readouts, audiobooks, podcast intros, or scripts and tools. Its local, offline setup also suits users who want voice features without an account; cloud and mobile options are not yet available.

Voicebox in depth

Read the full write-up

Voicebox is a local AI voice studio for desktop users who want to create, edit, or route speech without an account. Its strongest case is an unusually broad free offline workflow; its main drawback is that cloud sync and mobile use are still forthcoming.

Overview

Voicebox brings voice cloning and speech generation together with a narrative editor, audio effects, dictation, dubbing, and output for compatible agents. It is open source, works offline, and supports commercial use. The breadth makes it useful to creators producing dialogue or other spoken material, as well as developers building voice into apps and agent workflows.

A clone can start from three seconds of audio, captured by upload, microphone, or system audio. The maker describes uses including game dialogue, audiobooks, podcast intros, accessibility readouts, apps, and scripts. Voicebox fits naturally among Text-to-speech tools, AI voice cloning tools, and AI voice changers.

Key features

Generation and story editing

Seven text-to-speech engines cover cloning and preset voices, with controls over delivery. That range gives creators options for different voices and performances without limiting the app to a single generation mode. The timeline-based Stories editor arranges multiple voices and tracks, trims clips, and mixes conversations, making it more suited to assembling a dialogue sequence than generating isolated lines alone.

Effects, dictation, and agents

Pitch shift, reverb, delay, compression, saved presets, and live preview provide practical sound-shaping tools. Dictation runs from a global keyboard shortcut and sends the transcript to the focused text field or clipboard; local Whisper models support 99 languages. This is a useful extension beyond voice production, particularly for people who want speech input without changing applications.

MCP-aware agents such as Claude Code, Cursor, and Cline can speak through Voicebox using a cloned voice. A POST /speak endpoint is described for other clients, while the local REST API also exposes speech generation, voice profiles, model status, generation history, and health checks. Developers get more than a voice interface: they can integrate and monitor generation locally.

Privacy and cloud direction

The desktop app works offline without an account, and its privacy policy says recordings, profiles, transcripts, generations, and settings are not sent to the maker. That makes the local workflow the clearest choice for users who want voice data to stay on their machine. The announced cloud backup encrypts library data on the user’s device, with servers storing ciphertext they cannot decrypt, but cloud service and mobile sync remain forthcoming.

Pricing

Local: free, free forever. It includes the full app, requires no account, and allows unlimited local generations and captures. This is the strongest option for individual creators and developers who can work on a supported desktop and do not need cloud synchronization.

Cloud: $12 a year at the launch price, coming soon. It is planned to include 25 GB of encrypted storage, up to 5 devices, 30-day version history, and encrypted backup and sync. The price and limits are not final, so this is not yet a dependable basis for choosing Voicebox or budgeting a workflow.

Studio: $48 a year is a placeholder, with pricing still to be determined. Its planned 250 GB storage, unlimited devices, one-year version history, and priority support target heavier multi-device use, but the unsettled terms make it premature to compare with established subscriptions. There is no free trial; the free Local plan already provides the full app rather than a time-limited evaluation.

Platforms

Desktop downloads are listed for macOS Apple Silicon and Intel, Windows 64-bit, and Linux builds from source. The broader platform list also includes Android, iOS, API, and self-hosted, but mobile and cloud sync are described as coming soon. Docker deployment is documented. In practice, the present appeal is strongest for desktop users; readers who need a ready mobile workflow should look elsewhere for now.

Who it's for

Voicebox suits creators making multi-voice narratives, game dialogue, audiobooks, or podcast intros, and developers who want local speech generation, agent output, or an API. Dictation and accessibility readouts broaden its usefulness beyond production work. It is less suitable for someone who depends on live cloud collaboration or mobile access today, or who needs settled cloud pricing and service limits.

Pros and cons

Pros: The free Local plan includes the full app and unlimited local generations and captures, so there is no usage cap on the core offline workflow.

Pros: Cloning, seven generation engines, a multi-voice timeline editor, effects, dictation, and agent output cover substantially different voice tasks in one desktop tool.

Pros: Offline operation without an account, alongside the stated local-data privacy policy, suits privacy-conscious users.

Cons: Cloud backup and mobile sync are still forthcoming, limiting users who need those workflows now.

Cons: Cloud and Studio pricing and limits are not final, making paid service costs and terms uncertain.

Cons: Linux installation is a build-from-source route, a less direct option than the listed packaged macOS and Windows downloads.

Alternatives

Choose Typecast if you want a freemium option with a free trial, Android, iOS, web, and API access; its Free plan is limited to 3 projects, 3,000 lifetime credits, 1 GB of media storage, and 720p video exports.

Google Cloud Text-to-Speech is a better fit for API- or web-based speech generation billed by character: Standard voices include 0 to 4 million characters free, then cost US$0.000004 per character (US$4 per 1 million characters).

Consider Fish Audio if its free tier’s 8,000 monthly credits and generation allowance suit your needs; the tier caps generations at 7 minutes and 500 characters per generation.

NaturalReader is another freemium option, with Android, iOS, browser extension, desktop, and web support.

Choose Speechify for a freemium tool spanning mobile, web, desktop, and API, with 10 voices and a maximum speed of 1.5x on its Free plan.

Altered Studio may suit someone seeking voice morphing, since its free plan includes 3 minutes per month, 10,000 AI Tokens, and local voice cloning; that plan has an attribution license and noncommercial-use restriction.

Consider LOVO AI for a freemium web and API option whose Free plan allows 5 projects.

GPT-SoVITS is a free alternative available across API, web, desktop, and self-hosted platforms.

Verdict

Voicebox is a strong choice for desktop creators and developers who want a free, open-source voice workflow that stays local and extends from cloning to editing, dictation, and agent speech. The unlimited Local plan and breadth of tools are compelling reasons to choose it. Look elsewhere if mobile or cloud sync is essential now, or if you need firm paid-plan terms before committing.

Verdict

Voicebox specifies a broad local voice workflow, from cloning and editing to dictation and agent output. The free Local plan is the clearest current option; cloud plans remain forthcoming with unsettled pricing and limits.

Where Specifiction places it

Alternatives to VoiceboxAll alternatives →