This project aims to transcribe the lyrics of a song from an audio file. The output file must have timestamps for each word/word part/phoneme. The more accurate the timestamps and transcription, the better the experience for the users trying to use the output for karaoke.
This project was done as part of the University of Texas at Austin's Case Studies in Machine Learning course.
- Use at least 30 sources, 15 of which must be academic papers
- Times new roman 12pt font
- Formatted minimum 10 page report without title/cover page, ToC, or references
- 1.5 line spacing
- Elaborate on
- Introduction and research background
- Research and methods
- Materials and Data sources
- Results
- Discussion and Conclusions
- Citations in APA format (Author, Year)