aws_polly_test.py

import boto3
import time
from datetime import timedelta
import sys
import os
import argparse


#We need to get our API credentials in the code for authentication that we have stored as Environment Variables locally.
os.environ.get("AWS_ACCESS_KEY_ID")
os.environ.get("AWS_SECRET_ACCESS_KEY")
os.environ.get("AWS_REGION")


#Following line is used to save all the console outputs in a text file.        
sys.stdout = open('polly_output.txt','a')

start_time = time.monotonic()


def input_file(text_file_path):
    global text
    if os.path.isfile(text_file_path):
        with open(text_file_path, 'r') as text_file:
            text = text_file.read()
    else:
        print("File doesn't exist in the directory!")


def polly_test():
    #Initialize amazon_polly client function
    polly_service = boto3.client(
        'polly',
    )
    
    #If we execute using full text input, the code throws Word Limit Exceeded Exception!
    my_text = text[0:300]    
    polly_response = polly_service.synthesize_speech(
        OutputFormat = 'mp3',
        Text = my_text,
        TextType = 'text',
        VoiceId = 'Joanna' #English - American female voice
    )

    #Following code block saves the audio stream as an mp3 file
    with open('polly_stream_test_text.mp3', 'wb') as audio_file:
        audio_file.write(polly_response['AudioStream'].read())


if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description = __doc__,
        formatter_class = argparse.RawDescriptionHelpFormatter)
    parser.add_argument(
        'text_file_path',
        help = 'The complete file path of the text file you want to convert from text to speech.')
    args = parser.parse_args()
    input_file(args.text_file_path)
    polly_test()


end_time = time.monotonic()
print("Execution_Time:", timedelta(seconds = end_time - start_time))