Aurable

• 

  Homepage of Aurable

• 

  Translation of the transcript

• 

  Summary and additional info options

Inspiration

Audio plays a vital role in our life. So why not analyze it?

What it does

Aurable is an app that provides the user with an analysis of the uploaded audio. It gives the transcript, translated transcript, summary, and allow user to export the transcript to PDF.

How we built it

For the analysis part, Deepgram plays a vital role. With its API, we can get the complete analysis like transcript, summary, words and topics from the uploaded audio. By using the streamlit, we can build the UI to interact with the user. For the translation part, itranslate package helps in providing the translation for multiple languages. Its better to have have the transcript in some paperback form right? Here comes the fpdf library which helps in PDF generation.

Tools used

• streamit - The fastest way to build and share data apps
• Deepgram - Python SDK for Deepgram's automated speech recognition APIs.
• itranslate - Google translate free and unlimited, itranslate since gtranslate is taken
• Gitpod - Automated dev environments in the cloud
• fpdf - Simple PDF generation

Challenges we ran into

• Challenges with JSON parsing
• Integrating with streamlit

Accomplishments that we're proud of

Better audio analysis

What we learned

• Learned about ASR

What's next for Aurable

• Live captioning conferences and creating a brief analysis on it.
• Verifying the authenticity of audios and videos

Built With

• deepgram
• python
• streamlit