VocaPoint

Inspiration

After witnessing the struggles of some of the presenters during the welcome event, we came up with a solution that would make presenting less cumbersome.

What it does

The user can upload their PDF files into the web app and use the app to present it. It uses an offline speech-detection and classification library to listen for keywords that the user can associate with each slide. Whenever a keyword is detected, the web app jumps to the corresponding slide.

How we built it

We used the Python micro-framework Flask to develop the web app. To make use of speech-detection functionality, PocketSphinx, a Python library, was used.

Challenges we ran into

The real-time streaming of audio files to be interpreted constantly, the fast response of the speech interpreter, inter-operability of the system on different platforms

Accomplishments that we're proud of

A working web application with a lot of potential. The main use case, the control of the slides via speech commands, was fulfilled.

What we learned

Platform-interoperability is a big hurdle that is easily underestimated, Googling is an essential skill, the APIs of the most important players are not always the best choice for your system

What's next for VocaPoint

Hopefully, in a few years, it will be much easier to build apps of this kind, thanks to the increasing popularity of voice technology. We would like to stay involved.

Built With

• css
• flask
• html
• javascript
• pocketsphinx
• python