Research Scientist 
​Speech, Google

​E-mail         Scholar         Linkedin

I am a researcher at Google, Speech group. My work explores topics in speech processing (speech synthesis, speaker adaptation, voice conversion, ASR, new applications), machine learning and signal processing. 
​

Prior to Google, I worked in Siri Apple from 2016. I was a PhD student in Center for Speech Technology Research (CSTR) at The University of Edinburgh (archived page), under the supervision of Junichi Yamagishi and Korin Richmond. I worked on designing vocoders and acoustic models for speech synthesis during the research in Speech Technology Group of Toshiba Research Europe from 2013 to 2015. I also worked on complex-valued neural network during my study in National Institute of Informatics (NII), Japan. Before working in Apple, I did the internship in BBC Research & Development, London.
Recent work:
A complete list of publications can be found here.

Whispered and Lombard Neural Speech Synthesis  [pdf] [samples]
Qiong Hu, Tobias Bleisch, Petko Petkov, Tuomo Raitio, Erik Marchi, Varun Lakshminarasimhan
SLT, 2021
Neural Text-to-Speech Adaptation from Low Quality Public Recordings  [pdf] [samples]
Qiong Hu, Erik Marchi, David Winarsky, Yannis Stylianou, Devang Naik, Sachin Kajarekar
SSW, 2019
Siri On-Device Deep Learning-Guided Unit Selection Text-to-Speech System  [pdf] [blog]
Tim Capes, Alistair Conkie, Ladan Golipour, Abie Hadjitarkhani, Qiong Hu, Nancy Huddleston, Jiangchuan Li, Matthias Neeracher, Kishore Prahallad, Tuomo Raitio, Ramya Rasipuram, Greg Townsend, David Winarsky, Zhizheng Wu, Hepeng Zhang
Interspeech, 2017
Statistical parametric speech synthesis based on sinusoidal models  [pdf]
Qiong Hu
PhD thesis, University of Edinburgh, 2016