How to Make a Simple Tensorflow Speech Recognizer

Share it with your friends Like

Thanks! Share it with your friends!


In this video, we’ll make a super simple speech recognizer in 20 lines of Python using the Tensorflow machine learning library. I go over the history of speech recognition research, then explain (and rap about) how we can build our own speech recognition system using the power of deep learning.

The code for this video is here:

Mick’s winning code:

The weekly challenge can be found at the end of the ‘Make a Game Bot’ video:

More learning resources:

Please subscribe! And like and comment. That’s what keeps me going.

And please support me on Patreon! I don’t work for anyone, although I did make a one-off video for OpenAI because I love them:
Follow me:
Facebook: Instagram: Instagram:
Signup for my newsletter for exciting updates in the field of AI:
Hit the Join button above to sign up to become a member of my channel for access to exclusive content!


W Bayne says:

Title is different than content…

Tom Lynn says:

You were merely annoying until you started rapping. Next!

Obed Campos Solano says:

hi. quick question: can the training be made through another dataset? i have a dataset made out of audios and i'd like to use it instead

Abdechafi Neji says:

raise URLError(err)

urllib.error.URLError: <urlopen error [Errno 11001] getaddrinfo failed> how can I solve this error??

B6 Infinity says:

pip install speech_data doesn't work!!! HELP!!!!

Sohail Hussain says:

how to make a front end to take user inputs

Toke Hiber says:

link to get data is broken. any suggestion?

PhilMad says:

Great show. Love it

aBraM says:

How does the training ever stop considering you've made an infinite loop? also after training how do I test it and give it an mp3 or other sound file to be translated to text?

Kevin Strieter says:

Does this work on Windows 10? I'm getting all kinds of errors. Undoubtedly it's something I did, but starting my trouble shooting with the operating system. Thanks.

Fatma Özüdoğru says:

The links in the file are not active. I can't reach the dataset. Would you activate the links? Thank you.

Jairo Martínez says:

Hello dear friend. I am carrying out a project that requires hardware that is capable of speech recognition of 15 monosyllables (which will always be the same) at a speed of recognition of each monosyllable between 5 and 15 milliseconds. After recognition, it must send a 4-bit data to a port where it knows which of the monosyllables has been recognized. I wish I could buy that development. Thanks in advance.

White Is The New Green says:

With a different presenter, this could have been moderately useful…

Ujjwal Khannal says:

Sir need your help on this topic need to talk with you

Ix Suomi says:

You should do Google's speaker events, their spokes person doesn't handle things right.

Mostak Ahmed says:

You are an living AI

prathamesh jadhav says:

on which ide did you run the program

Michael Nguyen says:

Can you make one on a plagarism recognizer?

Kapil Gund says:

speech_data is not found

Atharva Athaley says:

Can you make a video on Audio feature extraction. Like MFCC and spectrogram. There are no tutorials or guidelines for librosa

Matt Graves says:

Freaking awesome video brother! You sir are one of the world's gems.
Thank you for sharing your thoughts and profound computer science insight, with an amazingly written rap no less!

Falcon Lin says:

I’m impressed 😂😂

Write a comment