Eric Lam
Portrait of Eric Lam

Curriculum vitae

Eric Lam.鍾浩霖 · Ho-Lam Chung

Ph.D. student, Graduate Institute of Electrical Engineering, National Taiwan University · Taipei, Taiwan

I am a Ph.D. student at National Taiwan University, advised by Prof. Hung-yi Lee in the Speech Processing and Machine Learning Lab. My research focuses on speech and audio processing and large language models: speech recognition and synthesis, neural audio codecs and tokenization, spoken language models, and the benchmarks that evaluate them. I lead OpenFormosa, an open AI initiative building models for Taiwan's language, voices, and culture. Previously, I was a research intern at Microsoft Research and at the Centre for Perceptual and Interactive Intelligence (CPII) at CUHK. BlueMagpie-TTS was selected as an outstanding paper at ISCSLP 2026, and my work has been supported by a Meta Audiobox Responsible Generation Grant. I also build independent apps and maintain open-source tools used by the community.

01

Education

  1. Ph.D. student in Electrical Engineering, National Taiwan University

    Graduate Institute of Electrical Engineering, M.S. / Ph.D. · Speech Processing and Machine Learning Lab · Advisor Prof. Hung-yi Lee

    Sep 2021 – Present

  2. B.S. in Computer Science and Engineering, National Chung Hsing University

    Sep 2018 – Jun 2021

02

Experience

  1. Project lead, OpenFormosa · Taiwan

    Open models for Taiwan's language, voices, and culture, from tokenization and language modeling to speech synthesis.

    2026 – Present

  2. Large-scale pretraining contributor, 27B-parameter language model training

    Launched and monitored multi-node training on an 8×12 H100 cluster scheduled with Slurm, and built the job scripting, log monitoring, and checkpoint-management workflow.

    2025 – Present

  3. Research intern, Microsoft Research · Redmond, WA, USA

    Researchy Questions: a dataset of multi-perspective, decompositional questions for deep research (SIGIR 2025).

    Jun 2023 – Sep 2023

  4. Research intern, Centre for Perceptual and Interactive Intelligence (CPII), The Chinese University of Hong Kong · Hong Kong

    Rare-word recognition for low-resource Cantonese ASR through homophone extension and unified writing (ISCSLP 2022).

    Jun 2022 – Sep 2022

03

Honors & grants

  1. Selected Outstanding Paper, ISCSLP 2026 Details

    BlueMagpie-TTS was one of 8 papers chosen from the 110 accepted papers for the Selected Outstanding Papers oral session.

    2026

  2. Meta Audiobox Responsible Generation Grant Details

    Selected by Meta FAIR in March 2024 for the NTU Speech Processing Laboratory proposal "Developing a Human-Centric Quality Metric for Synthetic Speech for RLHF": access to Audiobox models and code, and funding of up to US$50,000. One of ten grants in the program.

    2024

04

Publications

28 papers, grouped by topic and listed newest first. * Equal contribution. All papers on Google Scholar

Speech synthesis & audio generation6

  1. Listen, Critique, and Refine: RL-Based Self-Refinement for Instruction-Following Speech Synthesis

    Chee-En Yu, Yi-Cheng Lin, Sung-Feng Huang, Yun-Shao Tsai, Ho-Lam Chung, Xuanjun Chen, Hung-yi Lee

    arXiv preprint2026

    arXiv

  2. FdAudio: MeanFlow-Anchored Fréchet-Distance Post-Training for One-Step Text-to-Audio Generation

    Kuan-Po Huang, Bo-Ru Lu, Ho-Lam Chung, Shih-Hsin Wang, Hung-yi Lee

    arXiv preprint2026

    arXiv Project page

  3. BlueMagpie-TTS: A Token-Efficient Tokenizer, Language Model, and TTS for Taiwanese-Accent Code-Switching Speech

    Ho-Lam Chung, Bo-Xuan Zheng, Cheng-Chieh Huang, Cheng-Han Chang, Jung-Ching Chen, Lok-Lam Ieong, Ting-Lin Hsiao, Yu-Cheng Lee, Yi-Hsin Chung, Yu-Kai Guo, Hung-yi Lee

    ISCSLP2026Selected Outstanding Paper

    arXiv Demo Award

  4. Fréchet Distance Loss on Speech Representations for Text-to-Speech Synthesis

    Ho-Lam Chung, Kuan-Po Huang, Bo-Ru Lu, Hung-yi Lee

    arXiv preprint2026

    arXiv

  5. BreezyVoice: Adapting TTS for Taiwanese Mandarin with Enhanced Polyphone Disambiguation: Challenges and Insights

    Chan-Jan Hsu, Yi-Cheng Lin, Chia-Chun Lin, Wei-Chih Chen, Ho-Lam Chung, Chen-An Li, Yi-Chang Chen, Chien-Yu Yu, Ming-Ji Lee, Chien-Cheng Chen, Ru-Heng Huang, Hung-yi Lee, Da-Shan Shiu

    ISCSLP2026

    arXiv

  6. Towards General-Purpose Text-Instruction-Guided Voice Conversion

    Chun-Yi Kuan, Chen-An Li, Tsu-Yuan Hsu, Tse-Yang Lin, Ho-Lam Chung, Kai-Wei Chang, Shuo-Yiin Chang, Hung-yi Lee

    ASRU2023

    arXiv

Speech recognition5

  1. Context-Aware ASR for Mandarin Technical Lectures

    Ho-Lam Chung, Yiming Chen, Hung-yi Lee

    arXiv preprint2026

    arXiv

  2. Listen, Think, Transcribe: Continuous Latent Test-Time Scaling for ASR

    Ho-Lam Chung, Yiming Chen, Dau-Cheng Lyu, Hsiao-Tsung Hung, Hung-yi Lee

    ISCSLP2026

    arXiv

  3. A Self-Refining Framework for Enhancing ASR Using TTS-Synthesized Data

    Cheng-Kang Chou, Chan-Jan Hsu, Ho-Lam Chung, Liang-Hsuan Tseng, Hsi-Chun Cheng, Yu-Kuan Fu, Kuan-Po Huang, Hung-yi Lee

    ASRU2025

    arXiv

  4. ML-SUPERB: Multilingual Speech Universal PERformance Benchmark

    Jiatong Shi, Dan Berrebbi, William Chen, Ho-Lam Chung, En-Pei Hu, Wei-Ping Huang, Xuankai Chang, Shang-Wen Li, Abdelrahman Mohamed, Hung-yi Lee, Shinji Watanabe

    Interspeech2023

    arXiv

  5. Improving Rare Words Recognition through Homophone Extension and Unified Writing for Low-Resource Cantonese Speech Recognition

    Ho-Lam Chung, Junan Li, Pengfei Liu, Wai-Kim Leung, Xixin Wu, Helen Meng

    ISCSLP2022

    arXiv

Audio codecs, tokenization & evaluation4

  1. LLM-Codec: Neural Audio Codec Meets Language Model Objectives

    Ho-Lam Chung, Yiming Chen, Hung-yi Lee

    Findings of ACL2026

    ACL Anthology arXiv Code

  2. Codec-SUPERB @ SLT 2024: A Lightweight Benchmark for Neural Audio Codec Models

    Haibin Wu, Xuanjun Chen, Yi-Cheng Lin, Kai-Wei Chang, Jiawei Du, Ke-Han Lu, Alexander H. Liu, Ho-Lam Chung, Yuan-Kuei Wu, Dongchao Yang, Songxiang Liu, Yi-Chiao Wu, Xu Tan, James Glass, Shinji Watanabe, Hung-yi Lee

    SLT2024

    arXiv Challenge

  3. Codec-SUPERB: An In-Depth Analysis of Sound Codec Models

    Haibin Wu*, Ho-Lam Chung*, Yi-Cheng Lin, Yuan-Kuei Wu, Xuanjun Chen, Yu-Chi Pai, Hsiu-Hsuan Wang, Kai-Wei Chang, Alexander H. Liu, Hung-yi Lee

    Findings of ACL2024

    ACL Anthology arXiv Code Leaderboard

  4. Towards Audio Language Modeling: An Overview

    Haibin Wu, Xuanjun Chen, Yi-Cheng Lin, Kai-Wei Chang, Ho-Lam Chung, Alexander H. Liu, Hung-yi Lee

    arXiv preprint2024

    arXiv

Spoken language models & audio understanding8

  1. Escaping the Procrustean Bed: Groupwise Orthogonal Connectors for Audio-Language Models

    Ho-Lam Chung, Ke-Han Lu, Yi-Cheng Lin, Guan-Ting Lin, Yiming Chen, Hung-yi Lee

    arXiv preprint2026

    arXiv

  2. TAU: A Benchmark for Cultural Sound Understanding Beyond Semantics

    Yi-Cheng Lin, Yu-Hua Chen, Jia-Kai Dong, Yueh-Hsuan Huang, Szu-Chi Chen, Yu-Chen Chen, Chih-Yao Chen, Yu-Jung Lin, Yu-Ling Chen, Zih-Yu Chen, I-Ning Tsai, Hsiu-Hsuan Wang, Ho-Lam Chung, Ke-Han Lu, Hung-yi Lee

    ICASSP2026

    arXiv

  3. TW-Sound580K: Dual-ASR Verification and Arbitration for Localized Audio-Language Modeling

    Hao-Hui Xie, Ho-Lam Chung, Yi-Cheng Lin, Ke-Han Lu, Wenze Ren, Xie Chen, Hung-yi Lee

    ISCSLP2026

    arXiv

  4. Building a Taiwanese Mandarin Spoken Language Model: A First Attempt

    Chih-Kai Yang, Yu-Kuan Fu, Chen-An Li, Yi-Cheng Lin, Yu-Xiang Lin, Wei-Chih Chen, Ho-Lam Chung, Chun-Yi Kuan, Wei-Ping Huang, Ke-Han Lu, Tzu-Quan Lin, Hsiu-Hsuan Wang, En-Pei Hu, Chan-Jan Hsu, Liang-Hsuan Tseng, I-Hsiang Chiu, Ulin Sanga, Xuanjun Chen, Po-Chun Hsu, Shu-Wen Yang, Hung-yi Lee

    arXiv preprint2024

    arXiv

  5. GSQA: An End-to-End Model for Generative Spoken Question Answering

    Min-Han Shih, Ho-Lam Chung, Yu-Chi Pai, Ming-Hao Hsu, Guan-Ting Lin, Shang-Wen Li, Hung-yi Lee

    Interspeech2024

    arXiv

  6. Bridging Speech and Textual Pre-trained Models with Unsupervised ASR

    Jiatong Shi, Chan-Jan Hsu, Ho-Lam Chung, Dongji Gao, Paola Garcia, Shinji Watanabe, Ann Lee, Hung-yi Lee

    ICASSP2023

    IEEE Xplore arXiv

  7. T5lephone: Bridging Speech and Text Self-Supervised Models for Spoken Language Understanding via Phoneme-Level T5

    Chan-Jan Hsu, Ho-Lam Chung, Hung-yi Lee, Yu Tsao

    ICASSP2023

    IEEE Xplore arXiv

  8. DUAL: Discrete Spoken Unit Adaptive Learning for Textless Spoken Question Answering

    Guan-Ting Lin, Yung-Sung Chuang, Ho-Lam Chung, Shu-Wen Yang, Hsuan-Jui Chen, Shuyan Dong, Shang-Wen Li, Abdelrahman Mohamed, Hung-yi Lee, Lin-Shan Lee

    Interspeech2022

    arXiv

Language models, question generation & agents5

  1. Researchy Questions: A Dataset of Multi-Perspective, Decompositional Questions for Deep Research

    Corby Rosset, Ho-Lam Chung, Guanghui Qin, Ethan C. Chau, Zhuo Feng, Ahmed Awadallah, Jennifer Neville, Nikhil Rao

    SIGIR2025

    DOI arXiv

  2. Handover QG: Question Generation by Decoder Fusion and Reinforcement Learning

    Ho-Lam Chung, Ying-Hong Chan, Yao-Chung Fan

    IEEE/ACM TASLP2024

    IEEE Xplore

  3. Keyword Provision Question Generation for Facilitating Educational Reading Comprehension Preparation

    Ying-Hong Chan, Ho-Lam Chung, Yao-Chung Fan

    INLG2022

    ACL Anthology arXiv

  4. Misleading Inference Generation via Proximal Policy Optimization

    Hsien-Yung Peng, Ho-Lam Chung, Ying-Hong Chan, Yao-Chung Fan

    PAKDD2022

    DOI

  5. A BERT-Based Distractor Generation Scheme with Multi-Tasking and Negative Answer Training Strategies

    Ho-Lam Chung, Ying-Hong Chan, Yao-Chung Fan

    Findings of EMNLP2020

    ACL Anthology

05

Open source & models

Selected repositories, OpenFormosa models, and learning resources. Follow @voidful on GitHub

LLMs, RLHF & Agent Training

TextRLAwesome ChatGPT DatasetAwesome Agent DatasetTaiwan Agent Bench

Speech, Audio Codecs & Voice AI

Codec-SUPERBLLM-CodecFDSpeechSTITCH Agent

Research & AI Skills

Hung-Yi Lee SkillAcademic SkillsAlchemist PlaybookRefCheck

Taiwan Data & Analytics

Taiwan Stock ArchiveInstitutional StockerTaiwan Company Network

Developer Tools & Local-First Apps

GPU Info APIaidevDevLEGOShadowSpacePhotoSpaceGuard

OpenFormosa models

BarbetBlueMagpie-TTSPangolinTokenizer

Learning resource

LeeDL: Interactive Traditional Chinese Edition (20 chapters)

06

Independent apps

10 apps on the App Store for iPhone, iPad, Mac, and Apple Watch. See all apps on the App Store

07

Skills & languages

Speech processing
Multilingual speech recognition, speech synthesis, neural audio codecs, speech–text model alignment, textless NLP
Natural language processing
LLM training with reinforcement learning, question decomposition and generation, distractor generation, prompt generation and selection
Distributed training
Slurm, multi-node H100 training, checkpoint management
Development & systems
Python, PyTorch, Hugging Face Transformers, Kubernetes, Helm, Kubespray, Dagster, Elasticsearch, distributed data crawling, iOS and macOS app development
Languages
Mandarin Chinese (native) · Cantonese (native) · English (fluent)

Last updated November 2026