Speaker Recognition | By Award Winning Textbook Author (Udemy.com)

O programa do curso é excelente, mas estou usando recursos adicionais para compreender o curso, principalmente porque a transcrição/legenda do áudio é confusa e não reflete a fala do professor. Em um curso de análise de fala, esperava que as transcrições dos áudios das aulas estivessem em português ou ao menos corretas em inglês. A transcrição como está prejudica a compreensão. A qualidade do curso está sendo prejudicada pela legenda. Usando recursos externos para compreensão dos tópicos, o curso fica excelente.

Content is good and surprisingly thorough. There is a lot of to-the-point content, and a good amount of coding lessons. However, the coding lessons can be a little frustrating - things like having to reimplement numpy feels like a waste of time, and not being able to see/print a single test case are awkward roadblocks.

the course is very nicely structured , little bit more of coding exercises or a link to a site where we can have more problems inline with the lectures that would help. also the subtitle information is very poor, because of the accent of the instructor its slightly difficult to understand and when the subtitles are also wrong thre is a chance to loose context

The course is very well structured and ticks all the bases from a signals and communication point of view. Only thing is it is not meant for beginners, either in the field of signal processing or in the field on coding on python. Hence anyone taking this course should have an intermediate skill level in both python and signal processing else you will find the videos are not fully self explanatory and you will have to make use of a lot resources outside of Udemy to learn the concepts and basics

It's a really great course for advanced learners (python) or for the beginner who has a big desire to get new knowledge about this topic. Here you have to work hard and search for information. This course gives the direction, not all knowledge that makes this course really interesting and inspires for future investigation at this topic. Good quizzes, interesting assignments, and rare information, that is difficult to find elsewhere in one place, but here.

This tutorial discusses exactly what I was searching from a long time and got to know about topics I was missing in my research. Thanks, Quang for the unique content but looking forward to more hands-on practice in other tutorials if any are on the way.

As one of the leaders in his field he’s doing a great job dissecting the topic of speaker diarization and manages to teach it in an easy-to-follow and interesting, yet entertaining way. He’s also not simply teaching how to apply speaker diarization but also its history and background as well as outdated methods. I recommend this course to anyone who wants to get their feet wet in speaker diarization!

As far I learn from this course, actually totally good in speaker recognition, this course teach us from the bottom, but unfortunately I didn't see any single full code tutorial from initialization library, load data, until evaluate our speaker recognition model, I hope this course can attach at least on last session for anything we learn from this course. no need the expert one, just state-of the art model or i-vector is enough because we learn that model in this course, thank you.

This course is a great overview of speaker recognition. Looking at the overall workflow is very helpful. Definitely clearer than reading 20 years worth of papers! The instructor doesn't "pre-digest" big mathematical concepts for you, but that's OK. One question: why is there no mention of x-vectors from Povey at al.?

As one of many general audiences to this course and new to speaker recognition field, I found this course very easy to follow and interesting to watch. As Dr. Wang mentioned in the video, there is few good materials covering this topic nowadays. However, this course covers all important concepts step by step, starts from the fundamentals of acoustics, perception, audio processing, signal processing, and feature extraction, to the more advanced topics such as how to build speaker recognition systems. This course really opens the door for new joiners in this field. Dr. Wang also introduced different options for inference logic, loss function, neural network topologies, as well as many data processing techniques and machine learning approaches. Those topics are very useful for applications even beyond speaker recognition such as NLP app, recommendation system, image processing, computer visions, etc. I now have a big picture of the speaker recognition industry and my general knowledge in deep learing and data processing has been expanded. I highly recommend this course to people from all backgrounds.