Natural Voice Narration
Turns a script into natural-sounding narration in 90+ languages
by ElevenLabs
pyVideoTrans
by jianchang512AppMake video or audio
Translates a video into another language with new subtitles and an AI voice-over
The problem it solves
You made a video in Chinese and want an English version, but paying a translator and a voice actor for every video is out of reach.
For Video creatorsTeachers and course makers
- Before
- You transcribe, translate and type subtitles by hand, then record or buy a voice-over and line it up with the picture.
- After
- You pick the languages and voices, check the text at each step, and get a new video with translated subtitles and dubbing.
- Stars
- ★ 20k
- License
- GPL-3.0
- Cost
- Free
- Platforms
- Win · Mac · Linux
- Runs
- Local
Overview
pyVideoTrans is a free desktop program that turns a video in one language into a video in another. It listens to the speech, writes subtitles, translates them and dubs the result with AI voices, and you can pause at each step to fix the text. It also makes subtitles from audio, translates SRT files and turns text into speech, using local models or online services you choose. It suits creators, teachers and small teams who publish videos in more than one language.
It runs the whole chain from speech recognition to dubbing in one window, and free channels such as Edge-TTS let you start without paying for an API.
Try these first
Start with the first one: it is a small task you can check easily.
- Translate a one-minute clip from [source language] to [target language] with Google translation and an Edge-TTS voice, pausing after recognition to check the subtitles.First try · free channels
- Make an SRT subtitle file from [audio or video file] and tell the different speakers apart.
- Translate [subtitles.srt] into English without dubbing.
- Dub [subtitles.srt] with a voice cloned from [reference audio] using F5-TTS.
Example prompts written by Agent Flows Hub.
Features
From the creator's documentation.
| Feature | What it does | Runs |
|---|---|---|
| Full video translation | Speech recognition, subtitle translation, dubbing and video merge in one run | Your computer |
| Subtitles from audio | Batch-convert audio or video to SRT subtitles, with speaker separation | Your computer |
| Step-by-step editing | Pause after recognition, translation or dubbing to fix the text | Your computer |
| Voice cloning | Clone a voice with models such as F5-TTS, CosyVoice or GPT-SoVITS | Your computer |
| Online services | Optional channels such as DeepSeek, Gemini, OpenAI or Azure, with your own key | Online APIs |
| WebUI and command line | Run it in a browser on your network or from the command line | Your computer |
Good to know
You have a video in one language and want a version with translated subtitles and a dubbed voice. pyVideoTrans runs speech recognition, translation, dubbing and the final merge, and lets you correct the text between steps.
Limits
- The Windows package is a 2.7 GB archive: extract it to a folder path without spaces or Chinese characters and run sp.exe there, not from inside the archive. GPU speed-up needs an NVIDIA card with CUDA 12.8.
- The program is free, but paid channels such as DeepSeek or ElevenLabs bill you directly, and you remain responsible for the rights to any video you translate.
Notes by Agent Flows Hub (original project name: pyVideoTrans), written from the creator's documentation, not from running it ourselves. pyVideoTrans documentation: Windows download · pyVideoTrans README: features and source install
Stuck setting it up?
Tell us what happened. We take a look first; if it needs paid work, you get an estimate before anything starts.
More to make video or audio
Other apps
- Reactive Resume Builds a resume and cover letters in your browser, checks them and tracks job applications
- Read Frog Translates web pages and video subtitles side by side and saves new words as flashcards
- Screenpipe Records your screen and audio on your computer so you and your AI can search what you did
- Skill Recorder Records you doing a task once and turns it into a skill your AI agent can repeat
- SolidCraft Lets you or your AI agent design 3D parts you can print, on desktop or in the browser
- SurfSense Researches, summarises and edits your documents on your own computer, with no account