General UNIHIKER

How to use Azure Speech Recognition and Speech Synthesis

userHead nick.qq2023 2023-09-07 10:22:44 311 Views0 Replies

Official website link: https://azure.microsoft.com/en-us

Step 1: Account Login

Login with GitHub account

Login with email address

Login with phone or Skype

Click Next to complete the login

Step 2 Create resources

Step 3 Set parameters

Basics create

Subscription-> Default selection

Resource group-> Create a new resource group

Regin-> Select according to your region

Create a new project name

Pricing tier-> Depending on your choice of pricing

Then select Next.

Network selection

Then select Next.

Identity selection

Open or close according to your needs

Then select Next.

Tags choice

Default empty

Then select Next

Review + create

Review + Create

Then select Create.

Creation complete

Click on GO to resource

Get the key and Location/Region

Click the arrow to get the key and Location/Region

The Unihiker uses Azure services

Installation:

pip install azure.cognitiveservices.speech`

Usage example (code):

import time
import os
from azure.cognitiveservices.speech import  SpeechConfig

AZURE_SPEECH_KEY = "" # Fill key 
AZURE_SPEECH_REGION = "" # Enter Location/Region 


try:
    import azure.cognitiveservices.speech as speechsdk 
except ImportError:
    print("""
    Importing the Speech SDK for Python failed.
    Refer to
    https://docs.microsoft.com/azure/cognitive-services/speech-service/quickstart-python for
    installation instructions.
    """)
    import sys
    sys.exit(1)

# Set up the subscription info for the Speech Service:
# Replace with your own subscription key and service region (e.g., "japaneast").
speech_key, service_region = AZURE_SPEECH_KEY, AZURE_SPEECH_REGION

def tts(text):
    speech_config = SpeechConfig(subscription=speech_key, region=service_region)
    speech_config.speech_synthesis_language = "en-US" 
    speech_config.speech_synthesis_voice_name ="en-US-JennyNeural"
    # Creates a speech synthesizer for the specified language,
    # using the default speaker as audio output.
    speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config)
    # Receives a text from console input and synthesizes it to speaker.
    result = speech_synthesizer.speak_text_async(text).get()

def speech_recognize_once_from_mic():

    speech_config = speechsdk.SpeechConfig(subscription=speech_key, region=service_region, speech_recognition_language="en-US")

    audio_config = speechsdk.audio.AudioConfig(use_default_microphone=True)
    speech_recognizer = speechsdk.SpeechRecognizer(speech_config=speech_config, audio_config=audio_config)

    print("Speak into your microphone.")
    speech_recognition_result = speech_recognizer.recognize_once_async().get()

    if speech_recognition_result.reason == speechsdk.ResultReason.RecognizedSpeech:
        print("Recognized: {}".format(speech_recognition_result.text))
    elif speech_recognition_result.reason == speechsdk.ResultReason.NoMatch:
        print("No speech could be recognized: {}".format(speech_recognition_result.no_match_details))
    elif speech_recognition_result.reason == speechsdk.ResultReason.Canceled:
        cancellation_details = speech_recognition_result.cancellation_details
        print("Speech Recognition canceled: {}".format(cancellation_details.reason))
        if cancellation_details.reason == speechsdk.CancellationReason.Error:
            print("Error details: {}".format(cancellation_details.error_details))
            print("Did you set the speech resource key and region values?")

    tts(speech_recognition_result.text)

while True:
    speech_recognize_once_from_mic()