!! This won't run on insecure origins (http) because of the security problems with JavaScript's WebAudio API. To test the app, please run the following: python server.py
This is a webapp that uses wavenets - a text2speech model - that generates audio signals based on some text input from an image or video frame!
- Run the web app by
python run.py - Take a snap (this gets fed into the network)
- The Flask backend will do the work: convert pixels into text, text into generating audio signals
- Send audio signals back to front-end
- Enjoy!
- OCR (Optical Character Recognition) set up to translate any image to text
- Webcam enabled by getUserMedia to take input from your cam
- Flask server set up -> Ajax can now communicate between front and back-end
- Front-end done -> you can take a snap now and see what image will get fed into the network
####TODOs
- Save images to feed into model? Not sure what's up here just yet.
- Set up wavenets for generating audio signals
- Feed the text input into wavenets, and see what it generates (should be a human-like sound!)
- fix the ugly CSS
Ignore this line: python -mSimpleHTTPServer 8000 .