SpeechPal

Inspiration

People have lots of challenges with speech daily, and we needed to find a way to get feedback on speeches and conversations without having to spend lots of money on a speech therapist. We have also had many problems with making smooth presentations and wanted to find a solution to this without being too expensive or inaccessible.

What it does

Our software takes an input of the audio of the speech/presentation, looks at the tone and pitch, and offers personalized feedback based on your speech, telling the user how he talked vs how he should have talked for a certain type of speech scenario.

How we built it

To build this, we used many languages such as dart, python, and sql, and used a python voice analyzer library in order to make our code function. We maintained a database in Postgres in order to use with our code.

Challenges we ran into

One main challenge we ran into was the incorporation of the speech analyzer into the flutter app, which required many bypasses.

Accomplishments that we're proud of

One accomplishment that we are proud of is having a functional product while working in a short time frame, and having many issues with communication and collaboration, which had to be done over distance since we could not have many sessions in person together. We also solved a real world problem which affected each of us as people who had to present this project.

What we learned

We learned about the usage of the libraries PyAudio and SpeechRecognition, which allows the program to play audio and analyze it.

What's next for SpeechPal

SpeechPal can be improved and the app can be made with a better UI, along with more speech styles such as more formal debating. It can also be more accurate and work with different loudness of the input, based on the setting in which the presentation/debate is being made.