Yao-yuan Yang is a research scientist with 12 years of experience at the intersection of machine learning, speech/audio processing, and applied research, currently working on Gemini Data and conversational applications at Google DeepMind. He holds a PhD from UC San Diego and has a strong track record of translating research into production-ready tooling—contributions to torchaudio include integrating Tacotron2 and implementing an edit_distance Levenshtein function used in speech recognition pipelines. His background spans academia and industry, from EEG and active learning research at National Taiwan University to internships at Facebook (torchaudio) and Oath where he optimized video-ad pipelines. Known for building open-source software and reproducible research, he brings deep signal-processing expertise alongside practical engineering that accelerates model development and deployment.
Data manipulation and transformation for audio signal processing, powered by PyTorch
Role in this project:
ML Engineer
Contributions:148 reviews, 35 commits, 65 PRs in 4 months
Contributions summary:Yao-Yuan implemented and added the `edit_distance` function, which is used to compute the Levenshtein distance between two sequences, and integrated it within the speech recognition example of the `examples/pipeline_wav2letter` pipeline and test files. Additionally, the user added the Tacotron2 model and training script. This indicates a focus on integrating and developing models for audio processing and speech synthesis within the PyTorch ecosystem. The user's contributions leverage the PyTorch framework.
Contributions:7 commits, 12 pushes, 3 branches in 1 year 3 months
lookrobustnessdeep-learningcomputer-visioncloser
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.