Speech Enhancement Evaluation Using Speech Recognition algorithm

In the project we created a tool for evaluating speech denoising algorithms. The aim of the project is to evaluate speech enhancement algorithms using automatic speech recognition algorithms. The problem of evaluating the speech enhancement algorithm is mainly based on the difference between a noise-free signal and an intelligible signal for hearing. Since our goal is to test the quality of improvement in terms of human understanding, the choice of speech recognition algorithms is natural.
The project had three stages: 1) Getting familiar with the tools I would need to test differently: Baidu, OM-LSA, DTLN.2) Construction of an algorithm for evaluating speech enhancement algorithms 3) Comparing the speech enhancement algorithm to other speech quality metrics, such as MOS and testing the algorithm on files we recorded in the laboratory.
The project is based on a Linux system and consists of several parts: an algorithm for editing the text, an algorithm for editing wav files, algorithms for finding distance between text files, and a price function for calculating the error.

Speech Enhancement Evaluation Using Speech Recognition algorithm