Talk - GPT-2 meets Whisper in WebAssembly

Talk with an Artificial Intelligence in your browser. This demo uses:

• OpenAI's Whisper to listen to you as you speak in the microphone
• OpenAI's GPT-2 to generate text responses
• Web Speech API to vocalize the responses through your speakers

All of this runs locally in your browser using WebAssembly.
You can find more about this project on GitHub.

More examples: main | bench | stream | command | talk |

---

Select the models you would like to use and click the "Start" button to begin the conversation

---

Debug output:
Troubleshooting

The page does some heavy computations, so make sure:

• To use a modern web browser (e.g. Chrome, Firefox)
• To use a fast desktop or laptop computer (i.e. not a mobile phone)
• Your browser supports WASM Fixed-width SIMD

Note that these neural network models were not meant to be used in a browser, so the performance and
quality of the results may not be optimal. If you have any questions or suggestions, checkout the following discussion.

Here is a short video of the demo in action: https://youtu.be/LeWKl8t1-Hc