Speaker Recognition | By Award Winning Textbook Author
What you'll learn
- Basic concepts and core algorithms in speaker recognition
- Audio processing and acoustics
- Machine learning and deep learning basics
- Coding practice and toolkits for audio and speech
- Python and PyTorch for machine learning
- Building a speaker recognition system from scratch
- College level mathematics
- Experience with machine learning or coding will be a plus
This course is an introduction to speaker recognition techniques.
Speaker recognition lies in the intersection of audio processing, biometrics, and machine learning, and has various applications. You can find the application of speaker recognition on your smart phones, smart home devices, and various commercial services.
In this course, we will start with an introduction to the history of speaker recognition techniques, to see how it evolved from simple human efforts to modern deep learning based intelligent systems.
We will cover the basics of acoustics, perception, audio processing, signal processing, and feature extraction, so you don't need a background in these domains. We will also have an introduction of popular machine learning approaches, such as Gaussian mixture models, support vector machines, factor analysis, and neural networks.
We will focus on how to build speaker recognition systems based on acoustic features and machine learning models, with an emphasis on modern speaker recognition with deep learning, such as the different options for inference logic, loss function, and neural network topologies.
We will also talk about data processing techniques such as data cleansing, data augmentation, and data fusion.
We included lots of hands-on practices and coding examples for you to really master the topics introduced in this course, and a final project to guide you through building your own speaker recognition system from scratch.
If you are a college student interested in AI or signal processing, or a software engineer, system architect or product manager working with related technologies, then this course is definitely for you!
Who this course is for:
- College students or graduate students
- Engineers, researchers, and program managers in universities or industry
- General audience interested in AI
- Fans of cool technology
Dr. Quan Wang is currently a Staff Software Engineer at Google, managing the Speaker, Voice & Language team, and an IEEE Senior Member. He was a former Machine Learning Scientist at Amazon Alexa team. Quan had been leading the efforts to deploy advanced speaker recognition technologies to various products at Google, making Google Home the first smart home speaker to support multiple users in the market.
Quan has authored 50+ impactful patents and papers in speaker recognition, speaker diarization, voice separation, speech detection, language recognition and speech synthesis, with 2700+ citations. Quan's work has received coverage by top tech media including VentureBeat, TechCrunch, Engage and CNET.
Quan is the author of the textbook "Voice Identity Techniques: From core algorithms to engineering practice", which was selected by the bestselling books about AI leaderboard in China, and won the Distinguished Author of Year 2020 Award.