TTS Predicting emotion from text for TTS 정 성 희 2020. 10. 19. 11:56 Emotion label specified during synthesis No. Neutral Happy Sad Angry 1 2 3 4 Emotion is predicted from language model (no emotion supervision from human during synthesis stage) No. Neutral Happy Sad Angry 1 2 3 4 5 공유하기 게시글 관리 Sunghee's research blog 'TTS' 카테고리의 다른 글 Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search (0) 2021.03.31 Blizzard challenge 2020 (0) 2021.02.18 FastSpeech: Fast, Robust and Controllable Text to Speech (0) 2020.05.28 CHiVE: Varying prosody in speech synthesis with a linguistically driven dynamic hierarchical conditional variational network (0) 2020.05.25 Pitchtron: Towards audiobook generation from ordinary people’s voices (0) 2020.04.30 'TTS' Related Articles Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Search Blizzard challenge 2020 FastSpeech: Fast, Robust and Controllable Text to Speech CHiVE: Varying prosody in speech synthesis with a linguistically driven dynamic hierarchical conditional variational network
