This repo is an implementation for Arabic and English language speech trainer for babies from 2 years to 6 years To help them get rid of stuttering
- Python 3.x
- Dependencies: gtts, speechrecognition, nltk,sentence_transformers
- You can study the code to understand from this file
main.ipynb
Install the dependencies using pip
-
Prepare your input text file containing the Arabic or English sentences you want to compare. Each sentence should be on a new line, check
English_sentences.txt
file. -
Prepare the input records in a folder called
English_records
orArabic_records
contains the child's voice. -
Download
E_model.pkl
from here link and Import the model and load it from the PKL file:
# Load the model from the PKL file
with open('similarity_calculator.pkl', 'rb') as file:
model = pickle.load(file)
file_path = "path/to/your/text/file.txt"
folder_path = "path/to/your/folder"
average_similarity = model(file_path,folder_path)