far and away THE best explanation of HMM or any speech recognition related thing i have ever seen. love your technique. simple, visual, full explained. why can't the other youtube vids be this instructive?
Thank a ton gentlemen for making this informative video. Concepts like Hidden Markhov Model is quite hard to understand but after watching this video I am quite comfortable with the topic of HMM Kudos to you and your team.
Thank you so much for this!! I'm doing a project on speech recognition, and I was starting to wonder why I chose this topic because it's so overly complicated. But you guys made it easy to grasp and now I don't have to stress as much.
That's a great informative video of such a complex system. Thank you so much! I wonder where you got the likeliest probabilities of phonemes and vowels with the values, though. Is there such a thing?
NB: the probabilities used in the part of the HMM (e.g. 0,6; 0,9; ...) were just chosen to explain the concept and are not a true representation of the actual probabilities. You guys already wrote it! you're the best!
could you share a video that compares the performance of a markov chain on speech recognition and then a HMM doing speech recognition on the same speech
Thanks for watching and sharing your question! This video and our related knowledge was the result of a one-off workshop, so unfortunately our expertise in this matter doesn’t reach further than what we’ve put in the video. Good luck with your further research!
First of all, thanks for the interesting and good explanation. Is there any reference for this? Any papers or books where you get that information from? I need it for a scientific research. Would be nice if any one could help me out :)
Appreciate your nature of sharing your learning(s). Just curious to know if there are any good pointers to understand how HMM is trained for words with Phonemes. I mean how arrive at TranstionProbability matrix and Emission probability matrix. Is it through dictionary of words./Phonemes we train HMM ?
I'm sorry, I could make a guess, but I don't have an actual answer to that. This video is the result of a crash course we made as a peer-to-peer learning exercise, although this is not my field of expertise (at all). Since it has been some years now, I don't know/remember anything more than captured in this video. I do hope this video helped you in getting the basics, but for more advanced questions & information you'll have to look elsewhere. Good luck!
far and away THE best explanation of HMM or any speech recognition related thing i have ever seen. love your technique. simple, visual, full explained. why can't the other youtube vids be this instructive?
As a visual learner this is an amazing demonstration, thank you!!
This was BY FAR the best explanation I've seen on this topic. Thank you guys.
You guys broke that down so even I could see how the big picture works. Thank you so much....JT
Finally a much needed explanation.. As I've just started to get into a voice assistant from scratch in c++ lol 😂
Good luck in your journey. I'm trying the same thing.
I think this could be my first RUclips video where I got a clearest explanation of this speech recognition..hats off to you both. Thanks a ton!!!
this is the best explanation one can find on online platform. Much appreciated. Thankyou. May you rise and shine.
Best video so far for a non technical to understand speech recognition
Great video! Super easy to understand intro to HMM and NN. Exactly what I was looking for.
Let's complete the homework, for my midterm assignment !!!!
It was really cool way of expanation of HMM ... good one
well done guys the way you explained this makes totaly worth to watch
Thanks a lot for not just the initiative but the clarity of the explanation too
Thank a ton gentlemen for making this informative video.
Concepts like Hidden Markhov Model is quite hard to understand but after watching this video I am quite comfortable with the topic of HMM
Kudos to you and your team.
thank you soo much. such a great and clear explanation.
You guys have done a great job!
This is the best explanation of the underlining principle.
Thank you so much for this!! I'm doing a project on speech recognition, and I was starting to wonder why I chose this topic because it's so overly complicated. But you guys made it easy to grasp and now I don't have to stress as much.
To good. This is what I was searching for past 2 days.
Thanks ❤️
Excellent explanation!! Thankyou for simplifying such complicated topics
very nice and interesting! the person pretending the model is so cute!
Thank you!!
Wow, the explanation is very easy to understand and covers the topic in detail. Great job!!
Best video about Speech Recognition by far
thanks for the simple and short explanation
concise, simple, awesome! Thank you!!!
Wonderful explanation with great level of understanding. Thanks for this beautiful content.
I enjoyed it. You really simplified for me and I 'm thinking to go ahead and code my first speech recognition. Thanks fellas.
dude now I truly understand the concept. Liked it, keep going.
thank you guys for doing this video you have saved me by helping me to understand HMM and ANN for my presentation, thank you SO MUCH
Thank you so much for the detailed explanation!!! Great help for my research topic :)
Beautiful loved it,made it so simple
so far the best explanation. KUDOS!!!
thank you for this video....you have really done hard work to make this....i can clearly see that
best video on asr ever!😍
This is so well made, thank you guys!! :)
I wish you guys made more of these videos, this is great
Wow!! Very good communicatiors, great video, really helpful
excellent explanation! thank you.
Thank u so much for the video!! It was really helpfull!!
a great combo for sure...thanks a million.
Best one, thanks a lot.
Superb explanation
Exceptional
Brilliant 👏 👏 👏 👏 👏
Thank you so much 💓 💗 💛 💖
Amazing, deserves more likes/views
Great video, helped me out!
Nice job, thank you both, guys.
Excelente more than excellent!!!
great explanation... I was working on a project of speech recognition and couldnt understand much from reading...but this helped a lot :)
amazing presentation, and the info too, just love it.
Thanks alot man, you saved my day.
this video was so good
1:25 The video is supe helpful, just a tinay thought. I think calling analog to digital converter hardware would be appropriate .
Thanks a lot guys!!!
Good explanation .. Need more topics..plss
that was a great explanation. so helpful.
Nice job.
Thank you for such a wonderful explaination of such a tricky concept
can you also please provide the code for the same ?
Thank you! This was awesome
you should do more of this !
Thanks guys! Next for me is to check how deep learning is used for speech recognition
So good 👌🏻
That's a great informative video of such a complex system. Thank you so much! I wonder where you got the likeliest probabilities of phonemes and vowels with the values, though. Is there such a thing?
NB: the probabilities used in the part of the HMM (e.g. 0,6; 0,9; ...) were just chosen to explain the concept and are not a true representation of the actual probabilities.
You guys already wrote it! you're the best!
1:40 I see what you did there! ;)
Thanks people!!!
brilliant mate.cheers.can u make a video on tacotron ,concatenative..and all the major methods too.
Great Video. Can you please give an example of sequential nature of speech and in what terms HMM is not flexible?
could you share a video that compares the performance of a markov chain on speech recognition and then a HMM doing speech recognition on the same speech
Thanks for watching and sharing your question! This video and our related knowledge was the result of a one-off workshop, so unfortunately our expertise in this matter doesn’t reach further than what we’ve put in the video. Good luck with your further research!
Thanks for your explanation! Very helpful! What does the input look like for NN? How does NN and HMM work together?
Very good! Thank you guys.
Da you have a video about the GMM+HMM?
Many thanks for this introduction. Especially the HMM Model was perfect for me since I read about it and the formulas always got me confused.
Great stuff! Do you have any papers or resources to go from here?
Great explanation thanks
How is an hmm trained to define the good probabilities ?
good job
I love you guys
Sir what is the Data set for Speech To Text Machine learning Model?
First of all, thanks for the interesting and good explanation. Is there any reference for this? Any papers or books where you get that information from? I need it for a scientific research. Would be nice if any one could help me out :)
Did you find any?
Appreciate your nature of sharing your learning(s). Just curious to know if there are any good pointers to understand how HMM is trained for words with Phonemes. I mean how arrive at TranstionProbability matrix and Emission probability matrix. Is it through dictionary of words./Phonemes we train HMM ?
Hey, did you get any info about it? Was looking for the same thing
how to calculate the probabilities used in real application ? is it stored in a database or how ?
I'm sorry, I could make a guess, but I don't have an actual answer to that. This video is the result of a crash course we made as a peer-to-peer learning exercise, although this is not my field of expertise (at all). Since it has been some years now, I don't know/remember anything more than captured in this video. I do hope this video helped you in getting the basics, but for more advanced questions & information you'll have to look elsewhere. Good luck!
You could use a language model to calculate the probabilities.
Can you share your workshop?
06:43
Hola,hablo espàñol
For anyone wanting to know about Fourier transform can refer to these videos : ruclips.net/video/spUNpyF58BY/видео.html
well done guys the way you explained this makes totaly worth to watch