Inspiration
Wanted to create a fully conversational AI
What it does
It accesses your device's microphone to collect audio which is then rendered to text and back to speech to be played back to the user
How we built it
We used HTML, bootstrap, and javascript to collect the audio and Openai's whisper API to convert the given audio to text, and another API from ElevenLabs to get K-AI to talk back to the user
Challenges we ran into
- we got stuck for more than a few hours trying to get a working logic for our button to end audio recording had issues previously with its onclick event
- the program still isn't able to fully respond to the audio collected ## Accomplishments that we're proud of Getting as much work done as we did for such a massive project in the little time given. ## What we learned VS Code is annoying with how none of the necessary dependencies are pre-installed and the fact it heavily relies on extensions can lead to confusing and misleading errors when they're not installed. ## What's next for K-AI Getting K-AI to be fully functional
Built With
- bootstrap
- gpt3
- html5
- javascript
- openai
- whisper
Log in or sign up for Devpost to join the conversation.