Author

Gyeongmin Kim

h-index 4 View on Semantic Scholar

We have 1 of 20 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

Extracting Voice Styles from Frozen TTS Models via Gradient-Based Inverse Optimization

Some text-to-speech systems ship a synthesis model and preset style vectors but withhold the reference encoder that turns a recording into a style vector, so a user cannot obtain a style for a new voice by inverting the released pipeline with gradient descent.

Gyeongmin Kim · arXiv.org · 0 citations