Tien-hong Lo
Researcher at The Research Center for Psychological and Educational Testing
Taipei, Taiwan
Summary
👤
Senior🎓
Top SchoolTien-Hong Lo is a research engineer and Ph.D. candidate in Speech & Spoken Language Processing with 11 years of experience focused on Automatic Speech Recognition and computer-assisted language learning. Based in Taipei and jointly funded by NTNU and the Research Center for Psychological and Educational Testing, he blends academic research with applied R&D to improve ASR systems and mispronunciation detection for CALL. He contributes to major open-source toolkits like Kaldi and ESPnet, where his work spans neural-network decoding, score fusion, data pipelines, and build automation—practical contributions that help productionize speech models. Comfortable across ML engineering and DevOps tasks, he also explores spoken information retrieval, NLP, and deep learning, bringing a hands-on, toolkit-first approach to research problems.
11 years of coding experience
1 year of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at National Taiwan Normal UniversityBachelor, Computer Science, Bachelor, Computer Science at Fu Jen University
Github Skills (22)
pytorch10
python10
scripting10
command-line-interface10
machine-learning10
bash10
kaldi10
automation10
script10
deep-learning10
speech10
sh10
speech-recognition10
automations10
shell10
Programming languages (6)
Github contributions (5)
kaldi-asr/kaldi
Nov 2017 - Jul 2024
kaldi-asr/kaldi is the official location of the Kaldi project.
Role in this project:
ML Engineer
ML EngineerContributions:17 PRs, 6 comments in 6 years 9 months
Contributions summary:Tien-hong primarily contributes to the Kaldi project by modifying and enhancing scripts related to neural network processing. Their work focuses on improving the usability of decoding scripts, adding features to score fusion processes, and fixing bugs. The commits involve code modifications to accommodate the output of neural networks, aiming to improve the overall performance of speech recognition systems.
cudakaldiasrspeech-to-textkaldi-asr
espnet/espnet
Apr 2021 - Sep 2022
End-to-End Speech Processing Toolkit
Role in this project:DevOps Engineer & ML Engineer
DevOps Engineer & ML EngineerContributions:11 commits, 3 PRs, 8 comments in 1 year 5 months
Contributions summary:Tien-hong primarily focused on updating and maintaining scripts related to data processing and the build pipeline. They made modifications to scripts for downloading and untarring data, removing long/short data, and updating JSON configurations, suggesting involvement in data preparation for model training. Additionally, they merged code changes and updated dependencies related to beam search algorithms and other core components.
speech-recognitionspeech-separationchainerspoken-language-understandingspeech-processing
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.