Objective

This course is an in-depth exploration of audio processing using neural networks. Starting with an introduction to audio problems, the course covers a range of topics including audio features and human labels, filtering and digital signal processing for audio processing, audio and music tagging with convolutional neural networks, audio and music transcription with recurrent neural networks, audio compression, bridging audio and language with sequence-to-sequence models, symbolic music generation, audio and music generation with pipelines, vocoder, and autoregressive models, audio and music generation with VAEs and diffusion models, controllable audio and music generation from texts and multiple modalities, and open problems and future directions in the field.

這門課程探討音頻和音樂處理。这门课從音頻和音樂問題的介紹開始，涵蓋音頻特徵和人類標籤、數字信號處理、使用卷積神經網絡進行音頻和音樂標記、使用循環神經網絡進行音頻和音樂轉錄、音頻壓縮、使用序列到序列模型连接转录音頻。這門課程音頻探討和音樂生成：包括符號音樂生成、聲碼器、自回歸模型、使用VAE和擴散模型進行音頻和音樂生成、從文本和多種模式控制音頻和音樂生成。

Syllabus

1. Introduction: Audio and music problems.
2. Audio features and human labels.
3. Filtering, digital signal processing for audio processing (Assignment 1)
4. Audio and music tagging with convolutional neural networks.
5. Audio and music transcription with recurrent neural networks.
6. Audio compression.
7. Bridging audio and language: audio and music caption with sequence-to-sequence models (Assignment 2).
8. Symbolic music generation.
9. Audio and music generation: pipelines, vocoder, and autoregressive models.
10. Audio and music generation with probabilistic models, such as diffusion models.
11. Controllable audio and music generation from texts and multiple modalities (Assignment 3).
12. Open problems and future.

.

Learning Outcome

Through a combination of lectures, assignments, and projects, students will gain hands-on experience working with state-of-the-art tools and techniques for audio and music processing. By the end of the course, students will have a solid foundation in the latest techniques for audio and music processing using neural networks, and will be able to apply these techniques to real-world problems in the field.

Back to the List

Objective

With the rapid development and increasingly application of Artificial Intelligence (AI) technology, specific AI Integrated Circuits (IC) are required to provide powerful and efficient computing capability, which drives the needs in market for types of AI IC suitable for different application situation. Therefore, this course intends to provide the enrolled students with the typical technology, system architecture and design methodology for AI IC design. Moreover, software tool chains are also introduced. This course is to support and enable the students with improved capability to design AI IC.

The content of this course is composed of five sections: 1) Fundamentals of AI and typical algorithms; 2) Technology and system architecture of AI IC, as well as design methodology considering algorithm, software and hardware; 3) Software tool chains related to the application of AI IC; 4) Emerging semiconductor devices, computing modality, circuit design, system architecture, requirement and trend; 5) Practice of AI IC design. The students to be enrolled in this course are required to be with essential knowledge in AI algorithm, software and IC design.

人工智能技術的飛速發展和日益廣泛的應用，需要專門的人工智能芯片提供強大、高效的計算能力，因此形成了對適用不同應用場景的多型人工智能芯片的市場需求。有鑒於此，這門課程引導參加此課程的學生學習人工智能芯片的技術和系統架構、算法軟件硬件相結合的設計方法、以及相關的工具鏈，以使能學生提高設計人工智能芯片和相關軟件工具鏈的能力。

在本課程中學生將學習五部分的知識：1）人工智能基礎和代表性算法；2）人工智能芯片的技術基礎、芯片架構、算法軟件硬件相結合的設計方法；3）人工智能芯片相關的工具鏈；4）新興的半導體器件、計算方法、芯片設計、系統架構、需求與趨勢；5）人工智能芯片設計實踐。參加本課程的學生需要具有基本的人工智能算法知識、軟件知識和芯片設計知識。

Syllabus

Introduction to artificial intelligence IC
Fundamentals of AI algorithms
Characteristics of AI computing
Architectures of AI IC – Overview
Architectures of AI IC – Operator
Architectures of AI IC – Data flow
Architectures of AI IC – Advance
Design methodology of AI IC
Tool chains of AI IC
Emerging devices, architectures and trend

Learning Outcome

Upon successful completion of this course, students will be able to:
1. Understand the introductory knowledge and typical algorithms of AI computing.
2. Understand the typical techniques and representative architectures of AI IC.
3. Understand the Domain-Specific Architecture (DSA) design methodology including algorithm, hardware and software.
4. Understand the software tools for the application of AI IC.
5. Understand emerging semiconductor devices, system architecture, requirement and trend.
6. Obtain practical experience of system development including requirement analysis, algorithm optimization, IC architecture and module design.

Back to the List

Objective

RISC-V is the processor of choice after Intel x86 and ARM processor. RISC-V is an open core and is fully configurable to support a wide scope of applications with varying performance requirements. This course will cover basic processor design. RISC-V instruction set architecture will be described along with an elementary pipelined implementation. Based on the elementary design, different techniques to increase performance will be examined from different levels of cache to instruction level and data level parallelism. This course can be seen as a computer organization course based on RISC-V.

RISC-V 是繼 Intel x86 和 ARM 處理器之後的首選處理器。 RISC-V 是一個開放核心，完全可配置，可支援具有不同效能要求的廣泛應用。本課程將涵蓋基本的處理器設計。將描述 RISC-V 指令集架構以及基本的管線實作。基於基本設計，將從不同層級的快取到指令級和資料級並行性研究提高效能的不同技術。本課程可視為一門基於RISC-V的電腦組成課程。

Syllabus

Basic Computer Organization
RISC-V ISA
RISC-V Pipelined Datapath
Instruction Hazards
Memory Organization and Cache
Virtual Memory
Instruction Level Parallelism
Data Level Parallelism

Learning Outcome

1. Understand the basic concept of computer organization
2. Understand RISC-V ISA and appreciate the merit of different control implementations
3. Use memory organization to increase performance
4. Use different levels of parallelism to increase performance
5. Gain ability to configure a RISC-V

Back to the List

Objective

The emerging intelligence of automotive vehicles requires increasingly powerful computing capability embedded on vehicle, which motivates the eager demand for automotive Integrated Circuit (IC). Because of such specifics as high reliability, the automotive IC is quite distinct from consumer IC. This course intends to provide for the enrolled students with the knowledge of automotive IC, including the characteristics, fabrication process, development flow, testing, packaging, related software and application. The students can therefore be with fundamental knowledge and development capability of automotive IC.

The content of this course is composed of six parts: 1) Overview and particulars of automotive IC; 2) Fabrication process and devices of automotive IC; 3) Design of analog automotive IC; 4) Design of digital automotive IC; 5) Testing and packaging of automotive IC; 6) Software related to automotive IC. The students to be enrolled in this course are required to be with basic knowledge in physics, software and integrated circuits.

汽車的智能化需要日益強大的車載算力，從而激發了對汽車芯片的強烈需求。車載領域高可靠等的獨特性使得汽車芯片與消費電子芯片有很大的不同。本課程為參加此課程的學生提供汽車芯片的相關知識，包括汽車芯片的特點、製造工藝、開發流程、測試封裝、軟硬件結合以及應用。通過本課程學生將具有汽車芯片的基礎知識和基本的開發能力。

本課程的內容分為六部分：1）汽車芯片總覽與特性；2）汽車芯片製造工藝與器件；3）汽車芯片模擬電路開發；4）汽車芯片數字電路開發；5）汽車芯片測試與封裝；6）汽車芯片相關的軟件。參加本課程的學生需要具有基本的物理知識、軟件知識和芯片知識。

Syllabus

Introduction to automotive IC
Types and particulars of automotive IC
Fabrication process of automotive IC
Devices of automotive IC
Design of analog automotive IC
Design of digital automotive IC 
Testing and packaging of automotive IC 
Software related to automotive IC

Learning Outcome

Upon successful completion of this course, students will be able to:
1. Understand the basic knowledge and particulars of automotive IC. 
2. Understand the fabrication techniques and devices for automotive IC.
3. Understand the fundamental design techniques for analog and digital automotive IC.
4. Understand the knowledge about the testing and packaging of automotive IC.
5. Understand the software related to automotive IC.
6. Obtain practical experience of designing automotive IC. 

Back to the List