@bchodesigns: What if AI could bridge the gap between sign language and spoken English? This prototype uses MediaPipe to track 21 points on each hand and 468 points across the face in real time. Hand tracking recognizes the signs and builds them into a sentence while face tracking captures something more nuanced… emotion. American Sign Language and spoken English have fundamentally different grammar structures. In ASL, facial expressions don’t only show how you feel.. It can also completely change the meaning of what you’re saying. Once both the signs and emotions are captured, they get passed to a local LLM that interprets the emotional context and generates a natural English translation. The underlined words show exactly what the AI modified to make the sentence flow naturally. Finally, ElevenLabs brings it to life with voice. Tone is automatically adjusted to match the context and emotion. Used @Claude @ElevenLabs

bcho

Open In TikTok:

Region: US

Wednesday 15 April 2026 13:35:07 GMT