Guoguo Chen
Co-founder And CTO at Vobil.com
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolGuoguo Chen is a speech recognition expert and entrepreneur with 14 years of experience building production-grade voice systems and AI-driven contact-center and mobility applications. As co-founder and CTO of Seasalt.ai and Vobil.com, he combines product-forward startup leadership with hands-on engineering to ship cloud and on-device speech solutions. His career includes founding KITT.AI (acquired by Baidu), leading on-device speech and vision at Baidu DuerOS, and research contributions at Johns Hopkins that helped win IARPA’s ASpIRE challenge. A prolific open-source contributor, Guoguo has made substantive technical improvements to Kaldi and CNTK and maintained Snowboy’s cross-platform hotword detection code, including multi-format audio support and Android integration. He holds a PhD in speech recognition from Johns Hopkins and a bachelor’s from Tsinghua, and is known for turning academic advances in wake-word and sequence training into scalable, real-world products.
14 years of coding experience
10 years of employment as a software developer
Johns Hopkins UniversityBachelor's degree, Electronic Engineering, Bachelor's degree, Electronic Engineering at Tsinghua University
Github Skills (27)
c-language10
lib10
python10
scripting10
machine-learning10
audio-processing10
cntk10
kaldi10
script10
deep-learning10
tensorflow10
speech10
asr10
sh10
sequence-to-sequence10
Programming languages (8)
Github contributions (5)
Kitt-AI/snowboy
May 2016 - Mar 2020
Future versions with model training module will be maintained through a forked version here: https://github.com/seasalt-ai/snowboy
Role in this project:
Back-end Developer & Library Maintainer
Back-end Developer & Library MaintainerContributions:7 releases, 188 commits, 106 PRs in 3 years 10 months
Contributions summary:Guoguo primarily focused on enhancing and maintaining the `snowboy` library, updating its core functionality. They modified the library to return specific values for silence detection and extended `RunDetection()` to support various audio data formats, including string, float, int16, and int32. Furthermore, the user refactored and separated parts of the project to improve structure. The user's contributions extended to creating examples and fixing issues related to the C++ demo and Android library, which included the creation of the Android library.
pytorchsnowboydeep-learningmodel-trainingmachine-learning
athena-team/athena
Jan 2020 - Mar 2020
an open-source implementation of sequence-to-sequence based speech processing engine
Role in this project:ML Engineer
ML EngineerContributions:11 commits, 10 PRs, 11 pushes in 2 months
Contributions summary:Guoguo primarily focused on modifying and updating existing scripts and configuration files within the athena speech processing engine. They made changes to the data preparation pipeline, including modifying scripts for data conversion and transcript processing, and adjusted the training and decoding stages. The user also made modifications to the run scripts, moving configuration files and adjusting Horovod commands, which suggests involvement in model training and deployment using distributed training frameworks. Their work included changes to the model architecture, demonstrating an understanding of the underlying machine-learning models used in the project.
sequenceasrctcspeech-recognitionwfst
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.