@jainhm1227372

You're on a call with someone speaking Japanese or German, and you're both just... getting through it. Conduit translates your voice in real time during meetings, plays it back in your own cloned voice through a virtual microphone, and captions whatever the other person says — all on your laptop, no server, no subscription, no audio leaving your machine. Setup takes about 5 minutes. You open a full-screen enrollment UI, read complex sentences out loud, say "done" after each one. The system extracts a Fourier-domain fingerprint of your voice: your pitch, formant frequencies, the shape your vocal tract makes. That fingerprint is what keeps the output sounding like you rather than a generic TTS voice. During calls, the pipeline runs in sequence: Whisper transcribes your speech, NLLB-200 translates it across any of 8 languages (English, Spanish, Japanese, German, French, Italian, Russian, Hindi), and CosyVoice2 synthesises it in your cloned voice. That audio goes into a virtual microphone. Zoom and Teams see it as your mic. The other person hears you, in their language. Their side works in parallel. System audio gets captured, language auto-detected, translated, and shown as captions in an overlay that sits above your taskbar on Windows and above the dock on macOS — permanently, without blocking anything. The voice model gets better over time. When the GPU sits idle, a LoRA adapter trains quietly on your accumulated recordings, updating less than 0.5% of the TTS model's weights. After a few weeks it noticeably tightens. If local hardware isn't enough, training moves to Google Colab — you get a notebook, you run it, you paste the Drive link back. On an RTX 3050: ~3.6 GB VRAM total. Thermal monitoring cuts TTS if it runs hot. Once the models are downloaded, it works offline.

Build solutions that improve how software is built. Work with an AI that understands your codebase, helping you reduce repetitive work and build with real context. ⏳ Build your project in 48 hours. 🚀 Get hands-on with IBM Bob and explore how AI fits into real development workflows before it becomes standard. 💡 Build tools and workflows that developers would actually use. ⭐ Learn directly from experts throughout the event. 🤝 Join a team or build solo. 💰 Prize pool: $10,000 🧑🏻‍💻 Register before the Kick-Off Stream and start building with Bob.

Grab your ideas, spin up compute, and start building what’s next! Start with $100 in AMD Developer Cloud credits and begin building right away. Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further. To unlock credits and join the hackathon, sign up for the AMD AI Developer Program. 🤝 Go solo or team up — ship something real. 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs. 📍 On-site Venue (May 9-10): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 📍On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑‍💻 Developers, builders, and innovators - bring your boldest ideas and let’s turn them into real AI systems on AMD.