Mobile app voice assistant using Rasa

I want to develop a mobile app voice assistant using Rasa. I have seen blogs about Rasa integration with Mozilla Deepspeech to convert it to a voice assistant but I am assuming it works only for the desktop/server environment. Can someone please point me to something similar to Mozilla Deepspeech for the mobile environment (both iOS and Android)? Otherwise are there any pointers for mobile voice assistant/voice bot using Rasa?

Since Deepspeech is not a real time Speech2Text engine, it won’t be as intuitive as something like Google Assistant that gives real time text of your speech as you speak. Also putting speech to text model on a mobile app would be tough as the models have to be very concise. I suggest you host your bot servers separately and build an app that calls your bot APIs.

The better way to do it would be to use an existing bot framework like Google Assistant and link your Rasa bot with it. You can make use of Google Actions console.
Here is a great blog by @Juste from the Rasa team on integrating Rasa bot with Google Assistant: Going beyond hey google, building a rasa powered google assistant

1 Like

Thanks for the URL. Is there something similar for Siri, Alexa and Cortana. I am most interested in integrating with siri.

I’m not familiar with any SDK that is released for Siri development, then again I can be wrong since I’m not an iOS developer.

One way I think this can be done for iOS is that you build an app that calls your google action using APIs. Not sure of anything else apart from this.

For Alexa and Cortana you can make use of Dialogflow which opens up your google actions.

Select the Agent that you have created (the project that is mentioned in the google actions console) and then click on the fulfillment tab on the left.
You can see all the channels supported. just click on the channel you want like Alexa/Viber/Cortana/Slack and follow the steps mentioned there. It’s really simple to integrate Google Dialogflow/Actions to any bot platform that has open API. Everything is GUI based.