Inspiration

So with the rapid advancement of ai in recent days all organizations are very keen to use this new technologies to make its self stand out among its competitions and provide a better service to their use base so one of the important things to be considered to provide a good service is to solve problem of the customer and provide a quick solution to their problem. This can be done through chatbots which are widely available in the markets but this chatbots needs data to be trained upon based on organization specifics so my project helps the organizations to get their data. As most companies in India in earlier days and even now relies on phone conversations so as to solve a problem of the customer all the frequently asked question and answer data will be stored in the form call recordings so my project convert this audio file in text so that it can be used to train chatbots and fine tune the the chatbots as per requirements.

What it does

The app basically converts audio files to text file where the text file will be of the format of question and answers .This app can be used in diverse fields for example to build datasets, in courts to record proceedings ,to generate subtitles for a video, to summarize whole audio file within seconds and convert into text file etc

How we built it

this app is built using next js styling is done with tailwind and it uses hugging face api and open ai api as major ai tools to give the required output.

Challenges we ran into

As far as now there was no significant challenges.

Accomplishments that we're proud of

As a mechanical engineering graduate and transitioning from non-it to it this is my first hackathon which helped me learn many new things and I am proud of building my 1st app

What we learned

docker,hugging face models, open ai

What's next for Dataset Builder

my system doesnt support llm to be run on it so i couldnt integrate llm in the project. In future i would use cloud to run llms and fully build and deploy custom chatbots which can be used anywhere

Built With

Share this project:

Updates