Creating Voice-based Virtual Assistants Using NVIDIA Jarvis and Rasa

Originally published at: Creating Voice-based Virtual Assistants Using NVIDIA Jarvis and Rasa | NVIDIA Developer Blog

Virtual assistants have become part of our daily lives. We ask Siri almost anything that we wonder about or order groceries through Alexa. In addition to providing convenience to our daily lives, virtual assistants are of tremendous help when it comes to enterprise applications. For example, we use online chatbots to help navigate complex technical…

1 Like

Hi there! I’m following the weather chatbot demo from this link and I cannot seem to get microphone to work. The chatbot is only working via text (basically rasa capabilities with Jarvis UI). Any suggestions? I’m working on an EC2 instance powered by NVIDIA and running the UI locally from my Chrome browser. Thanks in advance!

Hi, were you able to install RASA on jetson nano? if yes then what version are you using?

Hi Saleem
We have not tried running RASA on jetson nano.
For our work, we ran it on the same hardware that we ran Jarvis on - a V100 GPU

Hi Hanna

Sorry for reaching out to so late. I seem to have missed this comment.

Were you able to get your mic working? Just want to make sure that you granted the website access to the mic.

Hi! The dialogue management component of the chatbot is developed using RASA. I was wondering if NVIDIA has this development tool too so that we can build a chatbot without relying on RASA?

1 Like

Hi Saeed
Thats a great question.
NVIDIA is currently working on working on a Chatbot Maker framework, built on top of Riva, which includes Dialog Management. This is currently under Early Access - You can apply for it here https://developer.nvidia.com/riva-studio-early-access

1 Like