Audio content is everywhere — podcasts, voice notes, video calls, interviews. But most transcription tools just give you raw text. No context. No emotion. No idea who's even speaking. EmotiCaption changes that. Powered by Deepgram, Hume AI, and a Hugging Face language model, EmotiCaption automatically separates speakers, detects genuine emotional expression in each phrase, and assigns contextually aware emojis — so you don't just read a conversation, you feel it. Whether you're going live with your mic, pasting a YouTube link, or uploading an MP3 or MP4 — EmotiCaption turns audio into a rich, accessible, emotionally intelligent transcript.
Built With
- deepgram
- huggingface
- hume
- react
Log in or sign up for Devpost to join the conversation.