Research Intern, Speech And Audio

Who Are We?
At HoYoverse, we are committed to creating immersive virtual world experiences for players around the world. In addition to game products such as Genshin Impact, Honkai Impact 3rd, Tears of Themis, and Honkai: Star Rail, HoYoverse also launched the dynamic desktop software N0va Desktop, the community product HoYoLAB, and created a variety of products such as animations, comics, music, novels, and merchandise around our original creative concept.
Adhering to our mission of Tech Otakus Save the World, we have always been committed to technology research and development, exploring cutting-edge technologies, and have accumulated leading technical capabilities in cartoon rendering, artificial intelligence, cloud gaming technology, and other fields.
HoYoverse is actively engaged in globalization, with offices in Singapore, Montreal, Los Angeles, Tokyo, Seoul, and other areas.
When you apply to a position with HoYoverse, we will process your personal data. To learn more about how we process your data, we encourage you to review our comprehensive Global Applicant and Candidate Privacy Policy . This policy provides detailed insights into how your information is collected, used, and protected throughout the application process.
Why Join us:
As a Research Intern for Lumi's expressive generative models, you will have opportunity to create better experience for users in the future.
T here will be both big challenges and opportunities ahead, thus we are looking for talented individuals to join our team as an intern in 2023/2024. This will be a hybrid role.
What You Will Do:

• Research, design, develop and evaluate generative models
• High-quality and expressive text-to-speech modeling
• Research to bring Large Language Modeling into Speech and Audio generation (Neutral Audio Codec, Audio Language Modeling)

What We Are Looking For:

• Familiar with the recent advance in the field of deep learning and generative models (Diffusion, AR, VAE, GAN, Flow, etc)
• First-author publication in top-tier conferences/journals (e.g., NeurIPS, ICML, ICLR, ICASSP, INTERSPEECH).
• Self-driven, innovative, collaborative, with good communication and presentation skills.

Do We Support Flexible Work?
Yes, as a remote team across North American time zones, we embrace flexible working. However, this role will be a hybrid role working remotely and in our US California office.
We are an equal opportunity employer that believes diverse backgrounds are key to bringing our concepts to life. If you're looking to play a key role in creating the best immersive virtual world experience for our users, we invite you to join our team.