I am currently pursuing my Ph.D. at the Machine Intelligence Lab, Seoul National University, under the guidance of Professor Kyomin Jung. My research has primarily focused on Text-to-Speech (TTS), particularly aligning text and speech modalities and separating linguistic information from other prosodic elements such as pitch and energy. My current research are centered on disentangling various speech components to enhance the performance and controllability of deep learning models across a range of speech processing tasks. In particular, I am interested in uncovering linguistic information in speech for Textless-NLP applications such as speech translation, an area I believe holds potential to bring people from all over the world closer together.
Education
Selected Publications
-
VarianceFlow: High-quality and Controllable Text-to-Speech
Using Variance Information via Normalizing Flow,
Yoonhyung Lee, Jinhyeok Yang, Kyomin Jung
ICASSP 2022
-
Bidirectional Variational Inference for Non-Autoregressive
Text-to-Speech. International Conference on Learning
Representation,
Yoonhyung Lee, Seunghyun Yoon, Kyomin Jung
ICLR 2021 [
Code ]
-
Multimodal Speech Emotion Recognition Using Cross Attention
with Aligned Audio and Text,
Yoonhyung Lee, Seunghyun Yoon, Kyomin Jung
InterSpeech 2020
-
Fast and Accurate Deep Bidirectional Language Representations
for Unsupervised Learning,
Joongbo Shin, Yoonhyung Lee, Seunghyun Yoon, Kyomin Jung
ACL 2020 [
Code ]
-
Effective Sentence Scoring Method Using BERT for Speech
Recognition,
Joongbo Shin, Yoonhyung Lee, Kyomin Jung
ACML 2019
-
MILAB at SemEval-2019 Task 3: Multi-View Turn-by-Turn Model
for Context-Aware Sentiment Analysis,
Yoonhyung Lee, Yanghoon Kim, Kyomin Jung
SemEval@NAACL-HLT 2019
Scholarships & Awards