Santanicoz
Posts: 4
Joined: Fri Jun 15, 2018 12:58 am

Comparing .wav audio

Tue Jun 19, 2018 1:46 am

Hi,

I would like to make a Isolated Word Recognition System that can recognize custom words. I've already made a simple python3 program that allows me to record multiple audio segments. Now I need to compare them and associate them with certain words so that the computer can recognize when I talk. Here is a video that explain exactly what I would like to do : https://www.youtube.com/watch?v=pnkUPFOw_Yo

The thing is that I don't know how to do that. Can anyone help me getting started with comparing and modulating sound waves so it can understand some words? Thank you for any help!

Santanicoz

User avatar
The Traveler
Posts: 361
Joined: Sat Oct 21, 2017 3:48 pm

Re: Comparing .wav audio

Tue Jun 19, 2018 2:35 am

Here's a start: Learn about Fast Fourier Tranform or FFT. In a nutshell, FFT allows one to analyze samples of a signal over time and divides it into it's component frequencies. Sounds, like words, have unique spectral signatures.

You take it from there.

Cheers.
Retired IT professional, C programmer and "beardie weirdie".
RPi interests: Developing an Infinite Improbability Drive
“Thinking outside of the box allows you to get rewards outside of your reach.” Matshona Dhliwayo

Santanicoz
Posts: 4
Joined: Fri Jun 15, 2018 12:58 am

Re: Comparing .wav audio

Tue Jun 19, 2018 2:56 am

Thanks, I will take a look.

Return to “Graphics, sound and multimedia”