A simple open source web interface for building voice assistants with Rasa. The interface records the voice input by connecting to the microphone enabled on your browser and autoplays the response generated by Rasa.
This is an alpha release of this voice interface. Community contributions are very welcome.
Necessary dependencies:
- node.js (https://nodejs.org)
- npm (https://www.npmjs.com/)
This is how you get the interface downloaded and ready to run:
git clone https://github.com/RasaHQ/rasa-voice-interface.git
cd rasa-voice-interface
npm install
And finally start the project with:
npm run serve
After that, the interface will start running on port 8080 and listen to port 5005. Launch the voice interface by pointing your browser to https://localhost:8080.
You can use this interface with any ASR systems you like. We tested this interface by integrating it with Mozilla DeepSpeech and Mozilla TTS components. A step-by-step implementation of this implementation is documented here.
To connect your own assistant to Rasa voice interface, you will need a connector. You can find an
example implementation of a custom connector in this tutorial. To adapt the interface to your
project, you can change the welcome message by editing the App.vue
and store.js
files.
Developed by INTEGR8 dev team, maintained by Rasa.
Licensed under the Apache License, Version 2.0. Copyright 2019 Rasa Technologies GmbH. Copy of the license.