🚀 Ultra-Fast Shuka v1 ASR

Optimized for speed • Multilingual • 15-second max clips

🎙️ Audio Input

🌍 Language Hint

Optional - helps with accuracy

📝 Transcription

⚡ Speed Optimizations Active:

✅ Auto audio trimming (15s max)
✅ CPU-optimized inference
✅ Minimal token generation
✅ Memory cleanup after each request

Tip: For fastest results, use short, clear audio clips in WAV format.