Open source alternatives to ElevenLabs

7 projects on starwire do the same job. Ranked by how much of what ElevenLabs does each one covers, read from its README and website. Ties go to the most active.

What ElevenLabs does

  1. Text to speech with realistic voicescore
  2. Voice cloningcore
  3. Dubbing and translation of audio
  4. Speech to text transcription
  5. Sound effects generation
  6. Voice library
  7. API for developers

Each dot in the table is one job, in this order. Square dots are core jobs and count double. Full means yes, half means partly. "See the evidence" under a project shows the line behind each one. Coverage is what a project documents, not how well it works or how hard it is to move to. How we score this

ProjectCoversStarsCommits, 12 weeksLast releaseClosed vs openedSetupLicense
speech-studioOffline Windows voice studio for text-to-speech, voice cloning and transcription Replaces it94%285+41
684
4 days ago89% Install like an app Custom license
Covers all 2 core jobs. Partly: Dubbing and translation of audio.See the evidence
  • Text to speech with realistic voicescoreYesvoice any text with cloned and built-in voices
  • Voice cloningcoreYesvoice any text with cloned and built-in voices
  • Dubbing and translation of audioPartlyA voice-over for a video
  • Speech to text transcriptionYesturn recordings into text with the speakers
  • Sound effects generationYesHiggs Audio v3 (cloning, emotions and sounds)
  • Voice libraryYesVoice library** — a voice pack in one click
  • API for developersYesspeaks the OpenAI audio API on the same port
yovoiceOffline audio studio for text-to-speech, voice cloning and songs Same job, different way89%2.2k+73
755
3 days ago93% Install like an app Apache-2.0
Covers all 2 core jobs. Partly: Dubbing and translation of audio; Voice library.See the evidence
  • Text to speech with realistic voicescoreYesTurn text into expressive speech
  • Voice cloningcoreYesclone a reference voice, design a voice from a description
  • Dubbing and translation of audioPartlyStory dubbing — import text or subtitle files
  • Speech to text transcriptionYesturn audio or video into timed subtitle lines
  • Sound effects generationYesinstrumental music and sound effects from an English description
  • Voice libraryPartlyBuilt-in voices | Kokoro-82M, Qwen3-TTS CustomVoice
  • API for developersYesAPI / MCP Deployment
VoiceStudioLocal voice cloning, dubbing, dictation and transcription app Replaces it78%111k+6.2k
5,999
2 days ago98% Install like an app AGPL-3.0
Covers all 2 core jobs. Not shown: Sound effects generation; Voice library.See the evidence
  • Text to speech with realistic voicescoreYesStories, audiobooks and queued jobs
  • Voice cloningcoreYesTurn a clean reference recording into speech in a familiar voice
  • Dubbing and translation of audioYesTranslate and dub videos with speech timed to the original.
  • Speech to text transcriptionYesA floating dictation widget and local transcription
  • Sound effects generationNot shownNeither the README nor the website mentions it.
  • Voice libraryNot shownNeither the README nor the website mentions it.
  • API for developersYesLocal API, MCP, reusable recipes and watch folders
voiceboxVoice studio for cloning voices, dictation and generating speech Replaces it78%61k+558
175
7 months ago100% Install like an app MIT
Covers all 2 core jobs. Not shown: Dubbing and translation of audio; Sound effects generation.See the evidence
  • Text to speech with realistic voicescoreYesgenerate speech in 23 languages across 8 TTS engines
  • Voice cloningcoreYesClone voices from a few seconds of audio
  • Dubbing and translation of audioNot shownNeither the README nor the website mentions it.
  • Speech to text transcriptionYesPowered by Whisper for accurate speech-to-text.
  • Sound effects generationNot shownNeither the README nor the website mentions it.
  • Voice libraryYes50+ curated preset voices via Kokoro and Qwen CustomVoice
  • API for developersYesREST API plus a built-in MCP server
soundboxLocal tool for sound effects, music, text to speech and voice cloning on a GPU Replaces it72%208+56
76
13 days agoToo few Install like an app Custom license
Covers all 2 core jobs. Partly: Voice library. Not shown: Dubbing and translation of audio; Speech to text transcription.See the evidence
  • Text to speech with realistic voicescoreYesText to speech from a short reference recording
  • Voice cloningcoreYesClone a voice from a short clip.
  • Dubbing and translation of audioNot shownNeither the README nor the website mentions it.
  • Speech to text transcriptionNot shownNeither the README nor the website mentions it.
  • Sound effects generationYesSound effects and ambience from a description
  • Voice libraryPartlyCast saved voices onto the lines of a script
  • API for developersYesA local HTTP API on port 17494
voice-proGradio UI for text-to-speech, voice cloning, transcription and vocal removal Covers one part72%13k+16
0
4 months agoToo few Needs a server GPL-3.0
Covers all 2 core jobs. Partly: Voice library. Not shown: Sound effects generation; API for developers.See the evidence
  • Text to speech with realistic voicescoreYesMultilingual text-to-speech
  • Voice cloningcoreYesZero-shot voice cloning
  • Dubbing and translation of audioYesmultilingual dubbing solution
  • Speech to text transcriptionYesTop-tier speech recognition
  • Sound effects generationNot shownNeither the README nor the website mentions it.
  • Voice libraryPartlyVoice samples for cloning will be continuously updated
  • API for developersNot shownNeither the README nor the website mentions it.
GPT-SoVITSFew-shot text-to-speech and voice cloning from about a minute of audio Covers one part56%65k+527
18
18 months ago41% Runs in a browser MIT
Covers all 2 core jobs. Partly: Dubbing and translation of audio; Speech to text transcription. Not shown: Sound effects generation; Voice library; API for developers.See the evidence
  • Text to speech with realistic voicescoreYesInput a 5-second vocal sample and experience instant text-to-speech
  • Voice cloningcoreYesfew shot voice cloning
  • Dubbing and translation of audioPartlyInference in languages different from the training dataset
  • Speech to text transcriptionPartlyChinese ASR, and text labeling
  • Sound effects generationNot shownNeither the README nor the website mentions it.
  • Voice libraryNot shownNeither the README nor the website mentions it.
  • API for developersNot shownNeither the README nor the website mentions it.

Vital signs from GitHub, checked weekly. How we read these · How setup is read

These also replace

All repos

New repos every two days, stars refreshed daily. Visits are counted without cookies, and the numbers are public. How it works Data RSS Newsletter

The best new open source, once a weekOne short email: the repo everyone is suddenly starring, a handful worth trying, and one hidden gem, each with a plain note on what it does. Free, with no tracking pixels.

Read past issues

↑ ↓ to move   Enter to choose   Esc to close

Follow starwire by RSSCopy an address into your feed reader.
The day's bestOne item a day: the repos rising fastest and the biggest new picks.starwire.foo/top.xml
Every new pickAll of them as they are added, up to 60 a day.starwire.foo/feed.xml
The weekly emailEach issue of the newsletter, once a week.list.starwire.foo/archive.xml