Voicebox is a free, open-source AI voice studio for your own computer. It clones a voice from seconds of audio, generates speech in 23 languages, and dictates anywhere, with no subscription.
Note: Clone only voices you have permission to use.
GitHub: jamiepine/voicebox
1 like · 29 views
1. 몇 초 목소리만 들려주면 그 목소리로 아무 글이나 읽어줘요.
전부 내 컴퓨터에서 돌아가서 목소리 자료가 밖으로 나가지 않아요.
jamiepine/voicebox
1 like · 2 views
어허.. TTS, AI 음성 생성 무료로 쓰는 방법 공유
얼마전 보이스박스라고 음성 복제 오픈소스로 나옴
jamiepine/voicebox
일레븐랩스나 피시오디오에서 고급-진 음성 생성하고
해당 녹음본을 그대로 클로드한테 복제하라고 시키는 것..
(단점: 괜찮은 GPU 필요, 클라우드 유료 서비스들보다 작업시간이 매우 오래 걸림)
미원엔비염n미혼앤비욘
32 likes · 12 reposts · 1,418 views
The Truth About VoiceStudio: Can It Replace ElevenLabs?
Looking for a free ElevenLabs alternative? We tried VoiceStudio, a free, open source app that clones a voice from a short clip, designs new voices from a description, dubs videos into other languages, transcribes, and turns scripts and books into audio, all on your own computer, with no account.
We cloned our narrator's voice from a twelve second clip, dubbed one of our own videos into Spanish, made a two-voice audiobook with chapters, then looked around ElevenLabs' own app to see what its plans give you for the money. And we found the catch: the app is free to use, even for work, but the voice engine it starts with is licensed for non-commercial use only, and nothing asked whose voice it was when we cloned one.
Tested on 26 October 2026: VoiceStudio v0.5.7, the official Docker image, on a PC with an NVIDIA RTX 5080 graphics card (16 GB), its interface recorded in Chrome on a Mac. ElevenLabs is shown in its own web app; we ran no tests on it. Prices are list prices in US dollars from ElevenLabs' pricing page on 26 October 2026: Starter $6 a month (the cheapest plan with voice cloning), Creator $22 a month; its first-month offers end on 1 November 2026.
What you need: Windows, a Mac or Linux. It's best with an NVIDIA graphics card or an Apple Silicon Mac, and it runs more slowly on a PC without one. The default voice model is about 3 GB, downloaded on first start. No account. Get it only from the project's releases page on GitHub (link below).
Project: debpalash/VoiceStudio
Should an app like this ask whose voice it is before it copies one? Tell us in the comments.
Chapters
0:00 Cold open
0:34 Setting it up
1:00 What it can do
4:20 Compared with ElevenLabs
5:35 The catch
7:23 Our verdict
Links
VoiceStudio's releases page, the only place to download it: github.com/debpalash/VoiceStudio/releases
ElevenLabs' plans and prices: elevenlabs.io
Voicebox, another free voice studio for Mac and Windows: jamiepine/voicebox
Download VoiceStudio only from its releases page on GitHub; fake fixes in its issues promise to unlock premium voices. The Windows installer isn't signed yet, so Windows shows "Windows protected your PC". Only clone a voice with its owner's permission, and for paid work don't use the default voice engine, which is for non-commercial use only.
VoiceStudio by Palash Debnath (debpalash), Shimul Debnath and contributors, maintained by Yupcha Softwares, AGPL-3.0 licence. Its default voice model, OmniVoice by k2-fsa, is licensed CC-BY-NC (non-commercial); VoxCPM2 by OpenBMB is Apache-2.0. ElevenLabs shown as the paid product we compared, from its own app and website; Voicebox by Jamie Pine (MIT) named as another free option. App icons are each product's own. Quotes from Palash Debnath's post "Why I Built an Open-Source ElevenLabs". Prices from ElevenLabs' US pricing page and voicestudio.sh/pro on 26 Oct 2026: ElevenLabs Starter $6 and Creator $22 a month, VoiceStudio Pro $99 a year. The video we dubbed and every voice we cloned are our own; the Alice in Wonderland passage is in the public domain.
#opensource #elevenlabs #voicecloning #aivoice #texttospeech #aidubbing #voicestudio #omnivoice
3 likes · 463 views
worth a run episode 01
Four AI tools. Are they worth your time?
In our first Worth a Run episode, we show what worked, what failed, and what still needs testing.
🎙 Voicebox produced a real short voice sample, but longer narration did not complete in our setup.
🖼 Upscayl enlarged a synthetic image from 256 × 256 to 1024 × 1024. We compare the same area before and after—more pixels do not necessarily mean recovered detail.
📄 AnythingLLM is an untested walkthrough. We did not verify an answer from our sample document.
🔗 Flowise launched locally, but its project is being discontinued. We would not choose it for a new production workflow.
Our approach: show the input, the result, and the limitation—then let you decide what is worth trying.
Chapters
0:00 Opening
0:03 How we test
0:16 Voicebox
0:51 Upscayl
1:26 AnythingLLM — untested walkthrough
1:48 Flowise — project status
2:12 Final verdict
Practical notes and sample materials on Telegram: [insert Telegram post link]
Projects:
Voicebox: jamiepine/voicebox
Upscayl: upscayl/upscayl
AnythingLLM: Mintplex-Labs/anything-llm
Flowise status: flowiseai.com
Subscribe to Worth a Run for useful finds, visible tests, and honest limitations.
#AITools #GitHub #WorthARun
3 likes · 49 views
@223Idrissa Voicebox nécessite au minimum 8 Go de RAM et 5 Go d’espace disque. Cependant, 16 Go de RAM et une carte graphique NVIDIA compatible CUDA sont recommandés pour faire tourner les modèles plus confortablement. Les besoins varient selon le modèle utilisé.
jamiepine/voicebox
3 likes · 1 repost · 152 views
那些要付费的语音工具，可以卸了——开源免费的 Voicebox 直接把 ElevenLabs 和 WisprFlow 一起端掉。
🔗 jamiepine/voicebox
能干啥？
1️⃣ 克隆任何人的声音
2️⃣ 全局语音输入随时用
3️⃣ 让你的 AI Agent 直接开口说话
最关键的是——100% 本地跑，不要 API Key，不限次数，数据压根不出你电脑。这种东西以前得花钱买，现在直接白嫖就行。
18 likes · 2 reposts · 778 views
Spacedrive: Your data should not have to live on somebody else's computer to be useful (39.1k stars)
What it does
All text below is quoted directly from the project README.
>Spacedrive is open personal data infrastructure. It turns the storage you control into one private filesystem across your devices, drives, servers, and the clouds you choose.
>Bring your data home without giving up search, protection, intelligence, sharing, or access from anywhere. No hosted account, no vendor control plane, no telemetry.
One filesystem, every source:
- A live index. Watchers keep names, sizes, timestamps, and content state current, so opening a folder never rescans the disk.
- Search across boundaries. One search runs across local, remote, cloud, and adapter sources.
- Stable identity. Records survive moves. BLAKE3 content identities connect copies of the same bytes across sources and devices.
- Fast duplicate detection. Sampled hashes find candidates quickly. Full-byte verification confirms equality before destructive cleanup.
- History that stays useful. Freeze a source at a point in time, browse an offline snapshot, and inspect which operation changed it.
- Provider exit. Move data away from a service without losing the organization, tags, history, and knowledge built around it.
>Spacedrive 2.0 Beta is planned for November 1, 2026.
Repository
spacedriveapp/spacedrive
Repository stats
|Metric|Value|
|:-|:-|
|Stars|39,094|
|Forks|1,354|
|Last release|0.4.3 (2025-03-24)|
|Languages|Rust 79%, TypeScript 18.1%, Python 1%|
|Contributors|130|
|AI contributor accounts (Claude / Codex / Cursor / Copilot)|claude (145 commits), cursoragent (83)|
|AI co-authored commits|Claude: 145|
Main contributor
|Field|Value|
|:-|:-|
|Name|Jamie Pine|
|GitHub|@jamiepine|
|Commits to this repo|2,218 (top contributor)|
|Bio|rust, ai and nice ui.|
|Company / Location|Spacedrive Technology Inc.|
|Website|voicebox.sh|
|X / Twitter|@jamiepine|
|Followers|2,804|
|Public repos|33|
|On GitHub since|2017-10-21|
----------------------------------------------------------------------------------
Discover the best GitHub repos without the endless scrolling.
Get top repos organized by category (AI, DevOps, Databases & more), plus key takeaways from developers' most useful comments - all in one free, twice-weekly newsletter.
👉 Get the best repos & community insights - Free
32 points · 1 comment
Mentions
Voicebox is a free, open-source AI voice studio for your own computer. It clones a voice from seconds of audio, generates speech in 23 languages, and dictates anywhere, with no subscription. Note: Clone only voices you have permission to use. GitHub: jamiepine/voicebox
1. 몇 초 목소리만 들려주면 그 목소리로 아무 글이나 읽어줘요. 전부 내 컴퓨터에서 돌아가서 목소리 자료가 밖으로 나가지 않아요. jamiepine/voicebox
어허.. TTS, AI 음성 생성 무료로 쓰는 방법 공유 얼마전 보이스박스라고 음성 복제 오픈소스로 나옴 jamiepine/voicebox 일레븐랩스나 피시오디오에서 고급-진 음성 생성하고 해당 녹음본을 그대로 클로드한테 복제하라고 시키는 것.. (단점: 괜찮은 GPU 필요, 클라우드 유료 서비스들보다 작업시간이 매우 오래 걸림) 미원엔비염n미혼앤비욘
The Truth About VoiceStudio: Can It Replace ElevenLabs? Looking for a free ElevenLabs alternative? We tried VoiceStudio, a free, open source app that clones a voice from a short clip, designs new voices from a description, dubs videos into other languages, transcribes, and turns scripts and books into audio, all on your own computer, with no account. We cloned our narrator's voice from a twelve second clip, dubbed one of our own videos into Spanish, made a two-voice audiobook with chapters, then looked around ElevenLabs' own app to see what its plans give you for the money. And we found the catch: the app is free to use, even for work, but the voice engine it starts with is licensed for non-commercial use only, and nothing asked whose voice it was when we cloned one. Tested on 26 October 2026: VoiceStudio v0.5.7, the official Docker image, on a PC with an NVIDIA RTX 5080 graphics card (16 GB), its interface recorded in Chrome on a Mac. ElevenLabs is shown in its own web app; we ran no tests on it. Prices are list prices in US dollars from ElevenLabs' pricing page on 26 October 2026: Starter $6 a month (the cheapest plan with voice cloning), Creator $22 a month; its first-month offers end on 1 November 2026. What you need: Windows, a Mac or Linux. It's best with an NVIDIA graphics card or an Apple Silicon Mac, and it runs more slowly on a PC without one. The default voice model is about 3 GB, downloaded on first start. No account. Get it only from the project's releases page on GitHub (link below). Project: debpalash/VoiceStudio Should an app like this ask whose voice it is before it copies one? Tell us in the comments. Chapters 0:00 Cold open 0:34 Setting it up 1:00 What it can do 4:20 Compared with ElevenLabs 5:35 The catch 7:23 Our verdict Links VoiceStudio's releases page, the only place to download it: github.com/debpalash/VoiceStudio/releases ElevenLabs' plans and prices: elevenlabs.io Voicebox, another free voice studio for Mac and Windows: jamiepine/voicebox Download VoiceStudio only from its releases page on GitHub; fake fixes in its issues promise to unlock premium voices. The Windows installer isn't signed yet, so Windows shows "Windows protected your PC". Only clone a voice with its owner's permission, and for paid work don't use the default voice engine, which is for non-commercial use only. VoiceStudio by Palash Debnath (debpalash), Shimul Debnath and contributors, maintained by Yupcha Softwares, AGPL-3.0 licence. Its default voice model, OmniVoice by k2-fsa, is licensed CC-BY-NC (non-commercial); VoxCPM2 by OpenBMB is Apache-2.0. ElevenLabs shown as the paid product we compared, from its own app and website; Voicebox by Jamie Pine (MIT) named as another free option. App icons are each product's own. Quotes from Palash Debnath's post "Why I Built an Open-Source ElevenLabs". Prices from ElevenLabs' US pricing page and voicestudio.sh/pro on 26 Oct 2026: ElevenLabs Starter $6 and Creator $22 a month, VoiceStudio Pro $99 a year. The video we dubbed and every voice we cloned are our own; the Alice in Wonderland passage is in the public domain. #opensource #elevenlabs #voicecloning #aivoice #texttospeech #aidubbing #voicestudio #omnivoice
worth a run episode 01 Four AI tools. Are they worth your time? In our first Worth a Run episode, we show what worked, what failed, and what still needs testing. 🎙 Voicebox produced a real short voice sample, but longer narration did not complete in our setup. 🖼 Upscayl enlarged a synthetic image from 256 × 256 to 1024 × 1024. We compare the same area before and after—more pixels do not necessarily mean recovered detail. 📄 AnythingLLM is an untested walkthrough. We did not verify an answer from our sample document. 🔗 Flowise launched locally, but its project is being discontinued. We would not choose it for a new production workflow. Our approach: show the input, the result, and the limitation—then let you decide what is worth trying. Chapters 0:00 Opening 0:03 How we test 0:16 Voicebox 0:51 Upscayl 1:26 AnythingLLM — untested walkthrough 1:48 Flowise — project status 2:12 Final verdict Practical notes and sample materials on Telegram: [insert Telegram post link] Projects: Voicebox: jamiepine/voicebox Upscayl: upscayl/upscayl AnythingLLM: Mintplex-Labs/anything-llm Flowise status: flowiseai.com Subscribe to Worth a Run for useful finds, visible tests, and honest limitations. #AITools #GitHub #WorthARun
@223Idrissa Voicebox nécessite au minimum 8 Go de RAM et 5 Go d’espace disque. Cependant, 16 Go de RAM et une carte graphique NVIDIA compatible CUDA sont recommandés pour faire tourner les modèles plus confortablement. Les besoins varient selon le modèle utilisé. jamiepine/voicebox
那些要付费的语音工具，可以卸了——开源免费的 Voicebox 直接把 ElevenLabs 和 WisprFlow 一起端掉。 🔗 jamiepine/voicebox 能干啥？ 1️⃣ 克隆任何人的声音 2️⃣ 全局语音输入随时用 3️⃣ 让你的 AI Agent 直接开口说话 最关键的是——100% 本地跑，不要 API Key，不限次数，数据压根不出你电脑。这种东西以前得花钱买，现在直接白嫖就行。
Spacedrive: Your data should not have to live on somebody else's computer to be useful (39.1k stars) What it does All text below is quoted directly from the project README. >Spacedrive is open personal data infrastructure. It turns the storage you control into one private filesystem across your devices, drives, servers, and the clouds you choose. >Bring your data home without giving up search, protection, intelligence, sharing, or access from anywhere. No hosted account, no vendor control plane, no telemetry. One filesystem, every source: - A live index. Watchers keep names, sizes, timestamps, and content state current, so opening a folder never rescans the disk. - Search across boundaries. One search runs across local, remote, cloud, and adapter sources. - Stable identity. Records survive moves. BLAKE3 content identities connect copies of the same bytes across sources and devices. - Fast duplicate detection. Sampled hashes find candidates quickly. Full-byte verification confirms equality before destructive cleanup. - History that stays useful. Freeze a source at a point in time, browse an offline snapshot, and inspect which operation changed it. - Provider exit. Move data away from a service without losing the organization, tags, history, and knowledge built around it. >Spacedrive 2.0 Beta is planned for November 1, 2026. Repository spacedriveapp/spacedrive Repository stats |Metric|Value| |:-|:-| |Stars|39,094| |Forks|1,354| |Last release|0.4.3 (2025-03-24)| |Languages|Rust 79%, TypeScript 18.1%, Python 1%| |Contributors|130| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|claude (145 commits), cursoragent (83)| |AI co-authored commits|Claude: 145| Main contributor |Field|Value| |:-|:-| |Name|Jamie Pine| |GitHub|@jamiepine| |Commits to this repo|2,218 (top contributor)| |Bio|rust, ai and nice ui.| |Company / Location|Spacedrive Technology Inc.| |Website|voicebox.sh| |X / Twitter|@jamiepine| |Followers|2,804| |Public repos|33| |On GitHub since|2017-10-21| ---------------------------------------------------------------------------------- Discover the best GitHub repos without the endless scrolling. Get top repos organized by category (AI, DevOps, Databases & more), plus key takeaways from developers' most useful comments - all in one free, twice-weekly newsletter. 👉 Get the best repos & community insights - Free