Wen-Chin Huang
@unilightwf
Followers
1K
Following
3K
Media
156
Statuses
2K
名古屋大学情報学研究科助教. Assistant professor, Nagoya University. Voice conversion & synthesis. Trilingual, street dancer, golfer. Tweets are my own opinions.
愛知 名古屋市 千種区
Joined June 2017
🚨I am honored to give an online invited talk at the Conversational AI Reading Group, MILA @convAI2024 on 5/15 11am-12pm EDT (5/16 0-1am Japan time), titled "Automatic Quality Assessment for Speech and Beyond"!.Please find more info on the website:
2
8
35
もちろん言いたいことわかるけど. 「ポスドク/助教として学ぶことがもうない=いる意味ない」ってことかな. 博士号を取った時点でもう指導教員/学生という関係性じゃなくなると思うから，もはや同僚.
0
0
1
RT @ShIma_Megagauss: 学位を取得した出身ラボでポスドク/助教をやる意味って何があるんやろ。.ボスからしたら手駒がいてありがたいやろけど、本人からしたらもう学ぶことはほとんどなくない？.
0
23
0
RT @CyberAgent_PR: AI Labの山本 克彦、村田 雅人ならびにリサーチインターンシップ参加者の党 少翔氏、He Jiajun氏、胡 鉦弘氏らによる論文6本が、音声言語処理分野の国際会議「INTERSPEECH 2025」に採択されました。. https://t….
cyberagent.co.jp
サイバーエージェントの新たな取り組みやサービス情報など、当社のニュースリリースを掲載しています。
0
16
0
$ whisper jp.wav.[00:00.000 --> 00:04.000] あらゆる現実をすべて自分の方へねじまけたのだ。. $ whisper --model medium --language Japanese --task translate jp.wav.[00:00.000 --> 00:04.000] I twisted all the reality towards myself.
0
0
1
RT @ArxivSound: Taejin Park, Ivan Medennikov, Kunal Dhawan, Weiqing Wang, He Huang, Nithin Rao Koluguri, Krishna C. Puvvada, Jagadeesh Bala….
arxiv.org
Sortformer is an encoder-based speaker diarization model designed for supervising speaker tagging in speech-to-text models. Instead of relying solely on permutation invariant loss (PIL),...
0
5
0
RT @welgene807: 今週の🔷中電シーティーアイ Welcome Generation🔷は….愛知県 #名古屋大学 からのインタビューでした🎙.来週の #ウェルジェネ 応援アーティストは #TaniYuuki さん🤗. WEB🖥.
0
10
0
I am invited to the APSIPA ASC 2025 Perspective & Panel session: "Neural Speech Assessment and Its Application". My presentation title: "Advancing Speech Quality Assessment Through Scientific Challenges and Open-source Activities". Looking forward to it!.
arxiv.org
Speech quality assessment (SQA) refers to the evaluation of speech quality, and developing an accurate automatic SQA method that reflects human perception has become increasingly important, in...
0
8
32
for each paper submitted you and your team are required to review 2 papers as a fair treatment to all authors and reviewers. マジ.
1
0
0
RT @erogol: KyutaiTTS solved streaming text-to-speech with a state machine that generates audio word-by-word as text arrives. 220ms latenc….
0
42
0
七月は.ASJ原稿執筆.APSIPA原稿執筆・学生原稿添削.初のASRU meta review（30本！）.数えないくらいの査読.今週の東海中間発表会の準備.今週と来週の講義準備.で消えた. 研究したいのぉ〜.
0
0
10
Completely no offense, but this is exactly how Japanese people pronounce "convolutional" 😂.
Cameron Churchwell, Minje Kim, Paris Smaragdis, "Combolutional Neural Networks,"
0
2
8
Yesssss.
Taesoo Kim, Jinju Kim, Dongchan Kim, Jong Hwan Ko, Gyeong-Moon Park, "Do Not Mimic My Voice: Speaker Identity Unlearning for Zero-Shot Text-to-Speech,"
0
0
5