Hi Product Hunt
I'm building Sokuji, an open-source real-time AI voice translation tool. Speak your language in a video call, and your words are translated and spoken in the other person's language live.
Why Sokuji?
Privacy-first, local-first.
Sokuji can run the entire speech recognition translation text-to-speech pipeline 100% locally in your browser using WebAssembly and WebGPU. No cloud, no API keys, no data leaving your device ever. Your conversations stay yours. Cloud providers (OpenAI, Gemini, etc.) are available as options, not requirements.
Just install a browser extension. That's it.
No desktop app download, no signup, no setup wizard. Install the Chrome/Edge extension, open your Google Meet / Teams / Zoom call, and start translating. It works in the browser sidebar right where your meeting already is. (A desktop app with system audio capture is also available for power users, but most people won't need it.)
---
What it can do
Offline local inference
- 12+ ASR models (SenseVoice, Whisper WebGPU, Paraformer, Parakeet TDT...)
- 49 Opus-MT translation pairs + Qwen 3.5 for multilingual translation
- Multiple TTS engines (Piper, Matcha, MeloTTS)
- Smart model variant selection based on your GPU capabilities
7+ cloud providers (optional)
OpenAI Realtime (GA, WebRTC), Google Gemini, Volcengine (ByteDance), Palabra AI, any OpenAI-compatible endpoint, or our hosted Kizuna AI option. Bring your own API key if you want lower latency or higher quality.
Modern audio pipeline
AudioWorklet-based low-latency processing, built-in noise suppression, dynamic device switching mid-session, cross-platform system audio capture, virtual audio devices on macOS/Windows.
35+ UI languages the app itself speaks your language too.
Other highlights: text input translation, karaoke-style TTS highlighting, Push-to-Talk, auto-updater (desktop), Stripe token wallet for hosted API.
---
What's next
The local-first approach is working, but we want to push it much further:
- TranslateGemma 4B Google's 55+ language any-to-any translation model via WebGPU, replacing current English-pivot Opus-MT limitation ([#123](https://github.com/kizuna-ai-lab...))
- Voxtral Mini 4B Mistral's 13-language real-time streaming ASR, <500ms latency ([#125](https://github.com/kizuna-ai-lab...))
- Native inference in Electron bypass WASM overhead, unlock GPU acceleration for significantly faster local performance ([#129](https://github.com/kizuna-ai-lab...))
- More meeting platforms: Webex, Jitsi, GoTo, RingCentral
- Windows & macOS code signing, Linux AppImage + Flatpak
---
Links
- Website: https://sokuji.kizuna.ai
- GitHub: https://github.com/kizuna-ai-lab...
- Chrome Web Store: https://chromewebstore.google.co...
- Desktop download (optional): https://github.com/kizuna-ai-lab...
Built by Kizuna AI Lab. Feedback, issues, and PRs welcome. If you deal with multilingual meetings or care about keeping your conversations private, I'd love to hear from you.

Here's something I noticed while building Auralify that I can't stop thinking about.

When a user asks ChatGPT, Meta AI, or Gemini about a company a brand reputation, a CEO, a product recall they get a confident, synthesized answer. No sources listed. No timestamp. No "this might be outdated." Just an answer. And users trust it. Recent studies suggest people now trust AI assistant responses more than the top organic search results.

Hey everyone,

Final year student here. Built Knot to solve my own problem - I had 2,000+ saved posts on X and remembered none of them.

What it does:

- Auto-saves tweets when you bookmark on X

Send bulk WhatsApp messages, automate campaigns, and track delivery, replies, and read status in real time.

Use your own number. No complexity.

Built for businesses that want to grow faster with WhatsApp.

Almost every founder I spoke to during research said the exact same thing. "We know our HR is a mess, we just haven't had time to fix it."

That one line changed how I'm building Office Bee.

The problem isn't that people don't know better tools exist. It's that switching feels like more work than just living with the spreadsheet. So nothing changes until something actually breaks.

We're building around that reality. Minimal setup, no long onboarding, just clean people ops that works from day one for startups that have outgrown spreadsheets but don't need a full enterprise system yet.

Hi guys,

For startup founders and venture capitalists, it is always difficult to see services' quantitative growth - mostly using excel to assume their future growth. However, DAU growth model based on your budget is so important to see long-term growth strategies. Without this, you might need to spend a lot of money which you wouldn't necessarily pay.

Social commerce will hit $2.6 trillion globally in 2026.
50.6% of US social media users made at least one social purchase last year.
Incredible scale.
But here's the uncomfortable question nobody asks:
Which brands are actually making profit from it?
TikTok Shop drives impulse buys.
Instagram drives aspiration clicks.
Both can destroy your margin if your unit economics are broken underneath.
Revenue through social is easy to celebrate.
Profit through social is something you have to deliberately engineer.
Knowing your real numbers per channel is not optional anymore - it's survival.

---------

#buildinpublic #analytics #data #shopify #ecommerce #saas #solofounder #okiela #shopifytips #ecommerceprofits #ecomtips #shopifytutorial #dropshippingtips #shopifystore #smallbusinessowner #aitools #profit #pricing

We have released the MVP version of Dageno AI.

Right now, the biggest challenge for us is that only a small number of users have tried the product so far. Because of that, we're not fully sure what the experience looks like from a user's perspective or what areas still need improvement.

Before our official launch, we want to make sure the product delivers a good experience, so we'd really appreciate any feedback, suggestions, or ideas from the community.