All of us have spent many minutes painfully searching through long videos to find a single sentence or phrase, and we know that many students share our struggle. We wanted to develop a program to do the work for us so that we can spend our time more productively.

What it does

cmd-f my voice. searches through your audio file to find the keywords that you specify and displays a list of timestamps so that you can easily skip to the most important parts.

How we built it

Combined HTML and Javascript to handle uploading file and displaying timestamps Used Google Cloud Storage to store user uploaded file Used Google Speech-to-Text API to transcribe the audio file from the cloud storage button and identify the timestamps of keywords. Created a node.js server to host the HTML/Javascript website which will be POSTing the obtained values from GCP Speech to Text API

Challenges we ran into

Learning on the fly, with three of us starting out with minimal exposure to JavaScript. Connecting backend and frontend.

Accomplishments that we're proud of

Collaborating to learn the necessary tools and overcome challenges. Compartmentalizing key tasks. Using an API!

What we learned

What an API is. How to use GitHub, git, and command line. The relationship between HTML and Javascript.

What's next for cmd-f my voice.

Optimize to receive larger audio files, video files. Minimize loss of quality in uploads to maximize accuracy of transcription. Augment to allow multiple keyword input.

Built With

Share this project: