Hey all 👋
If you are here, then you probably got inspired by the amazing work that Deepgram has done with it's Speech-to-Text service.
Here is a simple guide how to get started with this project and have yourself speaking into you browser in no time!
If for some reason you found this repository with no context, watch both of these videos in order, and you should be up to date. 😉
Create an account with Deepgram here, to get yourself $150 free credits (no credit card required). Once your account is created, generate yourself an API key. They have great instructions on their website which you can follow. Fork this repository and Clone it into you local folder on you machine. Add the newly generated API Key in the place of
DEEPGRAM_API_KEY
in the script.js
file.
Navigate to the project folder. Install all the dependancies by running npm install
After the dependancies have installed, you can run the project with npm run dev
to obtain your local IP address.
As the Media Capture and Streams API requres a secure HTTPS connection to interact with your microphone, you will need to install some SSL certificates for your project.
Generate the certificates by running:
mkcert YOUR_IP_HERE localhost
You can provide more than one IP address. Once those have been generated, you want to make sure they are placed in the
root
of your project.
Navigate to the
/bundler/webpack.dev.js
file, and replace YOUR_SSL_CERT_KEY
and YOUR_SSL_CERT
with a path to your certificate and key file generated in the previous step.
Make sure all your files have been saved and run npm run dev
again. Allow the browser to use your microphone.
Say something! Hurray! Your text should be visible on the screen! Now it's your time to experiment! Happy coding!!! 🚀