Skip to content
View leoohuang's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report leoohuang

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
leoohuang/README.md

Tianle Huang

AI Engineer  ·  MSc in AI & Language, Stockholm University

Typing SVG

LinkedIn Email


I design and build AI systems across the full research-to-deployment cycle. My work spans machine learning foundations, speech processing, and multilingual NLP — with a strong emphasis on rigorous evaluation, reproducibility, and practical deployment.


Competencies

Machine Learning

  • Statistical & representation learning
  • Classical ML methods
  • Experimental design
  • Evaluation methodology
  • Performance benchmarking

NLP

  • Multilingual embeddings
  • Cross-lingual semantic retrieval
  • Text preprocessing
  • Linguistic feature extraction
  • Generative system evaluation

Speech Technology

  • Automatic Speech Recognition
  • Whisper-based ASR pipelines
  • Speech signal analysis
  • Speech continuation experiments
  • Bias detection in ASR outputs

Stack

Python R Bash PyTorch HuggingFace Scikit-Learn NumPy Linux Docker Git


Projects

Multilingual Retrieval Engine  ·  Python PyTorch HuggingFace

Embedding-based semantic search system for cross-lingual similarity ranking.

  • Aligned multilingual vector representations
  • Cosine similarity scoring
  • Modular, reproducible evaluation pipeline
Speech Bias Evaluation Framework  ·  Python Whisper R

End-to-end pipeline for detecting and quantifying demographic bias in ASR outputs.

  • Automated Whisper transcription
  • Statistical significance testing (R + Python)
  • Reproducible experiment infrastructure
Human–LLM Speech Continuation Study  ·  Python Whisper

Experimental framework for benchmarking generative speech consistency against human baselines.

  • Audio segmentation and transcription workflow
  • Qualitative + quantitative evaluation design
  • Comparative analysis between human and model outputs

Research Interests

Generative AI evaluation  ·  Multilingual & cross-lingual NLP  ·  Speech–language integration  ·  Responsible AI

Stockholm, Sweden  ·  Open to research collaborations and engineering roles

Pinned Loading

  1. statistical-nlp statistical-nlp Public

    Python 1

  2. essay-score-regression essay-score-regression Public

    Python

  3. Podsearch Podsearch Public

    Tell me what topic you're interested in , and I'll find the exact podcast segments that actually discuss it.一个自动下载最新关注的播客+内容搜索系统。输入一句话,从播客音频中找到最相关的片段并直接播放。

    Python 3

  4. spoiler-detector spoiler-detector Public

    spoiler-detector for any film !

    Python 1

  5. maintenance-mind maintenance-mind Public

    LIS060 AI agent for self-reflective industrial maintenance retrieval

    Python

  6. obsidian-knowledge-ingest obsidian-knowledge-ingest Public

    Codex skill for turning source files into detailed, linked Obsidian knowledge domains

    Python