PolyTalk_BizzAppDev

joined 1 month ago
sorted by: new top controversial old
21
We built a privacy-first open-source tool for real-time speech-to-speech translation (lemmy.world)
 

For the past few days, we have been working on an open-source, self-hosted real-time speech-to-speech translation tool called PolyTalk. The goal was that there are people and organisations who need privacy around the tool they are using, and for the speech-to-speech translation, we haven't had many options.

We built the tool with Ollama, Faster Whisper, and Piper.

The tool is not limited to speech-to-speech translation only, but you can also share any of your tabs, whether you're watching a YouTube video in another language, the tool will give you audio output in your target language.

We are aware of how often context and tone get lost in translation, so we ensured translation quality by processing complete sentences instead of individual words.

Now we are focused on context support and tone adaptation.

If this interests you, here is the GitHub repo: https://github.com/PolyTalkIO/polytalk

Can Open-Source AI Bring Translation Back Under User Control? in c/opensource@lemmy.ml
[–] PolyTalk_BizzAppDev@lemmy.world 2 points 3 weeks ago

Fair point. I was trying to focus on the broader topic rather than lead with the project, but I can see why that might come across as marketing-style framing.

Can Open-Source AI Bring Translation Back Under User Control? in c/opensource@lemmy.ml
[–] PolyTalk_BizzAppDev@lemmy.world 1 points 3 weeks ago

That's a fair point. I think convenience will continue to win for a lot of people.

What interests me is having the option. For some use cases, a cloud service is perfectly fine. For others, whether it's privacy, compliance, reliability, or simply wanting control over your own infrastructure, self-hosted alternatives can be valuable even if they never become the default choice.

Also, the quality of open-source speech and translation tools has improved so much that they're becoming realistic options for far more people than they were a few years ago.

10
Can Open-Source AI Bring Translation Back Under User Control? (lemmy.world)
 

Most AI translation tools rely on cloud services.

Audio leaves your device, gets processed elsewhere, and comes back translated.

As open speech recognition, translation, and TTS models continue to improve, it feels increasingly possible to build communication tools that run on infrastructure users actually control.

That's one of the ideas behind PolyTalk, an open-source translation platform we're building.

Privacy, ownership, and transparency may soon matter as much as model quality.

Do you think communication tools like translation, transcription, and speech interfaces will eventually move back toward local and self-hosted deployments?

GitHub: https://github.com/PolyTalkIO/polytalk

1
What Does a Privacy-First AI Translation Stack Look Like? (github.com)
 

Most AI translation tools rely on cloud services.

Audio leaves your device, gets processed somewhere else, and comes back translated.

We wanted to explore a different approach.

PolyTalk is an open-source translation platform built around the idea that speech recognition, translation, and speech synthesis can be powered by open models and deployed on infrastructure you control.

The project combines open-source components for transcription, translation, and TTS into a privacy-first workflow.

Curious how others in the open-source AI community think about privacy and ownership when it comes to AI-powered communication tools.

GitHub: https://github.com/PolyTalkIO/polytalk

1
Built a privacy-first real-time translation platform with Ollama (github.com)
 

We've been building PolyTalk, an open-source real-time translation platform powered by Ollama.

Unlike most translation tools, it's not limited to speech-to-speech translation. It can translate audio from microphones, browser tabs, meetings, videos, and other audio sources in real time.

Current stack: • faster-whisper for speech-to-text • Ollama-compatible models for translation • Piper for text-to-speech

Privacy was a major goal, so the platform can run entirely on your own infrastructure.

Would love feedback from the community, especially around multilingual models and real-time translation workloads.

GitHub: https://github.com/PolyTalkIO/polytalk