About DubMochi
DubMochi began with a simple problem: a creator could be speaking to their viewers live, yet the conversation stopped when they didn’t share a language.
Why I built DubMochi
I wanted streamers to be able to speak in their own language and still be understood by the people watching. Putting the original words and a translation in the OBS scene lets viewers follow the conversation while the stream carries on.
Founder
SymphonyIceAttack · GitHubSymphonyIceAttack builds software in TypeScript and Go. Their public GitHub projects include a 3D 2048 game and a Go short-link platform.
What guides the design
One scene, two languages
A viewer watching a game or a drawing session needs to see the action as well as the words. The creator can place subtitles in an unused part of the scene and choose whether the audience sees the original, the translation, or both.
Caption style
Set the size and position of your captions. Show the original with the translation, or hide it to leave more space for your stream.
Inside DubMochi
In microphone mode, your browser detects speech locally. DubMochi sends audio through its Cloudflare Worker to Soniox for transcription and translation. The workspace saves final captions and their translations for export. Ended sessions and their captions are deleted after 30 days, and you can delete them sooner in the workspace.
DubMochi uses Google sign-in and Convex to manage your account, caption preferences and live session state.