A team of researchers from Johns Hopkins University has created new pitch correction technology called DiffPitcher. Unlike traditional pitch correction tools, such as Auto-Tune, DiffPitcher utilises AI to analyse and enhance vocal spectrograms, resulting in a more natural and higher-quality correction process.
According to Jiarui Hai, a PhD student involved in the project, "[the] results show that Diff-Pitcher outperforms previous pitch control methods in sound-quality and naturalness with great pitch controllability. Furthermore, we apply Diff-Pitcher in template-based and score-based automatic pitch correction systems and explore their application potentials."
The precision and control offered by DiffPitcher extend its applications beyond music, potentially revolutionising treatment for speech-related disorders such as post-laryngectomy rehabilitation and aiding stroke victims in voice recovery.
Presented at the 2023 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, DiffPitcher represents advanced pitch correction capabilities, namely the ability to maintain naturalness in corrected vocals while providing adjustable control over pitch marks a notable advancement in the field.