Translate
The speech is transcribed and translated into your chosen language.
Translate videos. Reach people.
Help more people understand your video. LinguoAI translates the speech and creates a new voice track with matching subtitles.
In your browser or on your computer. Your choice.
Schön, dass du da bist.
It's good to have you here.
Your content stays. The language changes.
A tutorial, an interview, or your next explainer video: no need to record it again.
The speech is transcribed and translated into your chosen language.
A new voice track is created and timed to match your video.
You also get the original and translated subtitles as SRT files.
The visuals stay the same. The new voice track replaces all original audio, including music and background sounds. Lip movements are not adjusted.
Your starting point
Three ways to get there. Choose the one that works for you.
Get started without installing anything locally. Run your own session on Google.
For regular use and control over your files. Processing runs on your computer.
For users who have received Studio credentials. Sign in and upload videos directly in your browser.
You can switch later. Download the project for the desktop app; open Colab and Video Studio using their links.
From your video to a new version
Load your video in the version you chose. A short clip is enough for your first try.
Choose the target language and a suitable voice. Start the translation.
Listen to the new version and check the subtitles. Save your video and the SRT files.
Also on your computer
For regular work and more control over your files. You need Python and FFmpeg; the guide walks you through setup.
Go to local installationSource code with the desktop app and Notebook. No ready-to-use installer: please follow the installation guide.
Download the source code ZIPGood to know
You can find technical details, voices, and advanced settings in the detailed guide.
The LinguoAI source code is freely available. External services and Colab plans may incur costs.
Google restricts the predominant use of external web interfaces in free Colab runtimes without a positive Compute unit balance. Check the Colab terms before starting.
On the desktop, you can use local models. In Colab, files are processed on a temporary Google VM; online translators and online voices send data to their provider. Only use content and voices for which you have the necessary rights.
No. Processing is also possible on the CPU, but takes longer. A short video is enough to get started. A GPU is optional.
Only while the corresponding Colab session and interface are running. The Gradio link is dynamic and should only be shared with people you trust. Download your results in time or save them to Google Drive.
Parakeet is an optional CPU speech recognition engine: NVIDIA v3 supports 25 European languages, while primeline specializes in German. Whisper remains the versatile default. Speed and quality depend on the recording and your computer.
Yes. Google Colab provides an interface for your active session. The separate Video Studio is in testing and requires credentials you have received; translation there is still limited.