- Home
- Text-to-speech tools
- VoiceStudio
VoiceStudio
What it is
VoiceStudio is an open-source, local voice AI studio for cloning and designing voices, dubbing video, dictating, transcribing, and creating audiobooks. To design a voice, users can describe traits such as age, accent, pitch, and emotion without providing a reference recording. Its video workflow transcribes and translates speech, revoices it, separates speakers, and aligns timing to the source. Scripts can become audio with multiple voices, while long text or EPUB files can be made into chaptered audiobooks. The desktop app provides a local OpenAI-compatible API for speech, voice, transcription, and dubbing functions. The maker lists 26 adapters, including local speech and transcription engines and a configured remote ASR adapter. Recordings, generated audio, transcripts, and derived voice data are stored on user-controlled storage, according to the maker; the app may still connect for updates, model downloads, or network-backed adapters. The public build has no hosted dashboard or Cloud API. Open Source is free under AGPL-3.0; Pro is $99 a year, and Lifetime is $299, paid once. Speech models have separate licences.
For and against
For
- Voice design does not require reference audio.
- Dubbing separates speakers and aligns timing.
- Audiobook creation accepts long text or EPUB.
- Exports include MP3, Opus, AAC, FLAC, WAV, and PCM.
Against
- Public build lacks a hosted dashboard and Cloud API.
- Model licences are separate; some may be research-only.
- Download FAQ recommends 16 GB or more RAM.
The verdict
VoiceStudio combines several voice workflows in a local studio, with a local API and multiple export formats. Check hardware needs and model licences before choosing it for a particular workload.
Who it suits
VoiceStudio suits people who want local voice creation, transcription, or dubbing tools across desktop and web platforms. Users considering commercial work should review the AGPL-3.0 terms and each speech model’s separate licence.
Plans and prices
Open Source
AGPL-3.0 · No seat count or evaluation period · Speech model licences are separate
Read 30 Sep 2026 · voicestudio.sh
Pro
1 user · 3 active devices · 3 concurrent uses
Read 30 Sep 2026 · voicestudio.sh
Lifetime
1 user · 3 active devices · 3 concurrent uses
Read 30 Sep 2026 · voicestudio.sh
Enterprise
Flexible terms for larger teams · Enquiry is not a purchase, quote, agreement, or licence grant
Read 30 Sep 2026 · voicestudio.sh
On the record
- API access
- Yesvoicestudio.sh · read 20 Sep 2026
- Audiobooks and stories
- Users can create multi-voice audio from scripts and chaptered audiobooks from long text or EPUB files.voicestudio.sh · read 30 Sep 2026
- Cloud status
- Cloud is in early access and invites users to request access for free usage credits; the public build does not offer the hosted dashboard or Cloud API.voicestudio.sh · read 30 Sep 2026
- Commercial use
- Yesvoicestudio.sh · read 20 Sep 2026
- Commercial use and licensing
- The maker says VoiceStudio may be used at work or for money under AGPL-3.0, while speech models have separate licences and some may be research-only.voicestudio.sh · read 30 Sep 2026
- Engine catalog
- The maker lists 26 adapters, including local text-to-speech and transcription engines plus a configured-remote OpenAI-compatible ASR adapter.voicestudio.sh · read 30 Sep 2026
- Export formats
- mp3,opus,aac,flac,wav,pcmvoicestudio.sh · read 20 Sep 2026
- Hardware
- The download FAQ says about 8 GB RAM and around 10 GB disk for models, recommends 16 GB or more RAM, and says GPU is optional but CPU use is slower.voicestudio.sh · read 30 Sep 2026
- Integrations
- The desktop app exposes an OpenAI-compatible local API at http://localhost:3900/v1 with speech, voices, transcription, and dubbing endpoints.voicestudio.sh · read 30 Sep 2026
- Maker
- The site says VoiceStudio was designed and built by Palash.dev, whose about page names Palash Debnath and describes him as based in Agartala, India; the opened pages give no founding year.palash.dev · read 30 Sep 2026
- Network behavior
- The app can access the network to check for updates, download models, or use an explicitly network-backed adapter.voicestudio.sh · read 30 Sep 2026
- Privacy
- The maker says local recordings, generated audio, transcripts, and derived voice data are stored on user-controlled storage and are not received by its website or control plane.voicestudio.sh · read 30 Sep 2026
- Product
- VoiceStudio is an open-source, local voice AI studio for voice cloning, voice design, video dubbing, dictation, transcription, and audiobook creation.voicestudio.sh · read 30 Sep 2026
- Video dubbing
- The dubbing workflow transcribes, translates, and re-voices video while keeping speakers separate and aligning timing with the original.voicestudio.sh · read 30 Sep 2026
- Voice cloning
- The desktop app clones a voice from a short clean audio clip, with about three seconds usually enough.voicestudio.sh · read 30 Sep 2026
- Voice design
- Users can describe a voice by gender, age, accent, pitch, and emotion in a sentence without providing reference audio.voicestudio.sh · read 30 Sep 2026
