I need a clean python code such that, could take take two audio files and compare them and print the pronunciation score in percentage showing how close they are.
Let’s say for an example, I need to test the word “chocolate.”
For the input arguments, the code should be able to in take perfectly pronounced audio file (.wav) and a recorded audio file (may in different length) of the same word and result a reasonable score showing how close those words are.
NOTE:
The code should remove noises and able to work in different audio lengths.
Comparing two waves should be same. And also should able to print a message if they are not literally same (Ex: if input audios are “chocolate” and “apple”).
Hello, Can you please confirm if you are looking at using any specific algorithms to do the analysis?
I will also provide documentation of code - the code will use certain open source python packages to achieve the comparison.
Already done projects related to voice analysis and signal recognition. It's all about the algorithm and the used approach witch you forgot to explain.