Inspiration
Optimization for life quality: we saw many people constantly looking at their phones to get public transport information and directions, since the voice instructions are either non-existing or not really like natural conversations. Statistics show that this has increased the cases of accidents. Additionally, there is no public facility for such voice assistants in the streets.
What it does
It is a natural language AI in combination with the city public transport info. In addition, the bigger concept is to implement such assistants in many locations in the city so that one may not necessarily need to use the phone to get the directions in the city
How we built it
Using NLP capabilities of Dialogflow to define intents and getting the transport information from Deutsche Bahn API. The intents then query through a webhook to python script to get conversational permutations.
Challenges we ran into
API connections, sometimes lack of API documentation or time limitation constraints due to document complexity
Accomplishments that we're proud of
Teamwork and that we managed to quickly assimilate new information from the workshops to implement in the product
What we learned
Lots of stuff about Dialogflow and speech recognition that we were unfamiliar with before the Hackathon
What's next for PT.ai
Improve the code and define the concept for the implementation in the pilot city
Built With
- dialogflow
- python
Log in or sign up for Devpost to join the conversation.