Slides
English slides, Mandarin delivery — 6 of 14 weeks ready
The slides are written in English to accompany the Mandarin lectures. Each deck is a reveal.js page — open it in a browser, nothing to install.
提示Keyboard shortcuts
B chalkboard layer for writing on a slide · ESC slide overview · F full screen · add ?print-pdf to the URL to print
| Week | Topic | Slides |
|---|---|---|
| W1 | A Systems View of Spoken Dialogue and the Latency Budget 語音對話系統的系統觀與延遲預算 |
▶ Open |
| W2 | Speech Signals, Auditory Front-Ends, and the Physics of Representation 語音訊號、聽覺前端與表徵的物理基礎 |
▶ Open |
| W3 | The Alignment Problem: From HMM to CTC 對齊問題：從 HMM 到 CTC |
▶ Open |
| W4 | Sequence Models and Streaming Architectures: Transformer, Conformer, RNN-T 序列模型與 streaming 架構：Transformer、Conformer、RNN-T |
▶ Open |
| W5 | Self-Supervised Representation Learning: Where Speech Foundation Models Come From 自監督表徵學習（SSL）：語音基礎模型的來源 |
▶ Open |
| W6 | Neural Audio Codecs and Discretization: Turning Speech into a Language Neural Audio Codec 與離散化：語音成為「語言」的關鍵一步 |
▶ Open |
| W7 | Speech Recognition: Three Paradigms, Streaming, and the LLM Turn 語音辨識：三種範式、streaming 與 LLM 化 |
Not yet available |
| W8 | Speech Synthesis I: Generative Paradigms and Streaming TTS 語音合成 I：生成範式與 streaming 合成 |
Not yet available |
| W9 | Speech Synthesis II: Controllability, Post-Training Alignment, Evaluation, and Misuse 語音合成 II：可控性、後訓練對齊、評估與濫用防範 |
Not yet available |
| W10 | The Physical Layer of Full-Duplex: VAD, AEC, Enhancement, Separation, Speaker 全雙工的「物理層」：VAD、AEC、增強、分離與說話人 |
Not yet available |
| W11 | Audio-Native Language Models: Building Half-Duplex Spoken Dialogue Audio-Native 語言模型：半雙工語音對話的建成 |
Not yet available |
| W12 | Full-Duplex I: Linguistic Foundations and Modeling of Turn-Taking 全雙工 I：Turn-Taking 的語言學基礎與建模 |
Not yet available |
| W13 | Full-Duplex II: Architectures, Multi-Stream Modeling, and Data 全雙工 II：架構路線、多串流建模與資料問題 |
Not yet available |
| W14 | Evaluation, Alignment, Deployment, and Open Problems 評估、對齊、落地與 Open Problems |
Not yet available |
The decks embed audio, so check that sound is on. Some clips are channel-separated (left = user, right = model); headphones or stereo speakers are recommended.
回到頂端