ibm_speech_api_stt.py

import json
from watson_developer_cloud import SpeechToTextV1
import time
from datetime import timedelta
import sys
import os
import io
import argparse


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

#Following line is used to save all the console output into a text file
sys.stdout = open('speech_api_content_stt_output.txt', 'w')

start_time = time.monotonic()


def input_file(speech_file_path):
	global result
	#Initialize SpeechToText function using the API credentials
	stt = SpeechToTextV1(
		username = STT_USER_WATSON,
		password = STT_PASS_WATSON)
	if os.path.isfile(speech_file_path):
		with io.open(speech_file_path, 'rb') as audio_file:
			result = stt.recognize(audio_file, content_type = 'audio/wav')  #This line will recognize the audio file and transcribe it into text format
	else:
		print("File doesn't exist in the directory!")


def speech_stt():
	text = result['results'][0]['alternatives'][0]['transcript']  #Text output of the audio file after transcription
	print("Text: " + text + "\n")


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

    input_file(args.speech_file_path)
    speech_stt()


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