This application was built with flask for python framework. It basically transcribes an audio file to its Text equivalent using IBM watson speech to text service.
- Bootstrap twitter framework was used for front end.
- The backend is done with Python using the Flask Framework
- Make sure you have an account on IBM cloud portal
- Make sure to have git set up on your local machine as described here.
- Once logged In, Click on Create link on the mainpage to create a speech-to-text resource. After which you can click on the show credential link to reveal the credentials necessary to use with this service.
- Make sure to copy api_key and url values into your config file or anywhere in your code.
- Copy the Values for api_key and url into
/instance/configfile of the flask project as shown below.
WATSON_API_KEY = 'your_api_key_value'
WATSON_API_URL = 'your_url_value'
- For this project do well to include the following code in the
/instance/configfile.
UPLOAD_FOLDER = os.path.dirname(os.path.abspath(__file__)) + '/uploads/'
- This line of code will serve as the url to be used to temporarily hold the uploaded audio file
- clone the github repo for this project by running the following commands in a terminal.
git clone https://github.com/nappy29/AudioToText.git
cd AudioToText
- It will be safer to run the app in development mode locally. So the following commands will be ok to have it running.
export FLASK_APP=audiotranscriptionapp
export FLASK_ENV=development
These commands are best suited for mac and Unix distro systems. If you running on a windows environment do well to
substitute export with set in the above commands
Enjoy :)