Part of PowerUpHacks Hackathon

Foreign Interviewees is purposed to help you gain every job opportunity possible. Our innovation will work past the bounds of language & will translate & speak for you in real-time, in your own voice!

With the interview season coming soon, everyone is boosting the resume, doing projects, practicing possible problems, and much more. In this chaos of things, adding another burden to be able to perfectly speak the interviewer's language is extremely daunting.

You may be the most qualified person, but without the confidence to speak about your achievements and respond clearly to the interviewer due to a language barrier, your chances decrease significantly.

My team member and I have seen this first hand with our family members and friends that live in other countries and wanted to take the initiative to help them out as much as we can.

Foreign Interviewees is a project purposed to help you, the interviewee, in effortlessly speaking any foreign language set out to you. Using real-time translations and deepfakes ML, our innovation allows you to comfortably speak in your native tongue while our system translates in real-time to the interviewer's language and speaks to the interviewer in your own voice!

Speech to Text Text to Speech Translation

Language Translator

Text to Speech through Voice Cloning

Language Translator

Text to Speech through Voice Cloning

• How to start
• Speech to Text coding
• Time Constraints
• Lack of Sleep :D

• Getting the translator to actually work
• Getting real-time audio recording and converting to different language on the spot
• This entire project is an accomplishment that we are both proud of
• The fact that everything works

Combination of Natural Language Processing & the opportunities and challenges it holds, as well as machine learning frameworks, networks, and techniques, including integrating APIs into our program.

Our next step is integrating our project with Google Meet, Zoom, and other meeting applications as an extension.

We will also take steps in making our tool sound more like the user: precising our deep learning framework in our implementation of the real-time Text-To-Speech Synthesis. We may try altering our Tacotron architecture or add more layers as to make our synthesized voice sound clearer and more like the individual.