Speech Recognition in Python using the SpeechRecognition Module

The SpeechRecognition module is a Python library that allows you to perform speech recognition tasks in your Python programs. It supports several speech engines, including Google’s Web Speech API, Google Cloud Speech API, Microsoft Bing Voice Recognition, and IBM Speech to Text. Here’s a tutorial on how to use the SpeechRecognition module in your Python programs:

Installing the SpeechRecognition module

Before you can use the SpeechRecognition module, you need to install it. You can install it using pip, the Python package manager. Open a terminal window and type the following command:


pip install SpeechRecognition

This will install the SpeechRecognition module and all its dependencies.

Using the SpeechRecognition module

To use the SpeechRecognition module in your Python program, you need to import it like this:


import speech_recognition as sr

Then, you can use the Recognizer class to create a recognizer object. The Recognizer class has several methods that allow you to perform speech recognition tasks, such as recognize_google(), recognize_bing(), and recognize_ibm().

Here’s an example of how to use the Recognizer class to recognize speech from a microphone:


import speech_recognition as sr

# Create a recognizer object
r = sr.Recognizer()

# Start listening to the microphone
with sr.Microphone() as source:
    print("Speak now: ")
    audio = r.listen(source)

# Recognize the speech
try:
    text = r.recognize_google(audio)
    print(f"You said: {text}")
except sr.UnknownValueError:
    print("Could not understand audio")
except sr.RequestError as e:
    print(f"Error making request: {e}")

In this example, the listen() method is used to listen to the microphone and record the audio. Then, the recognize_google() method is used to recognize the speech. If the speech is recognized successfully, the recognized text is printed to the console. If the speech is not recognized, an error message is printed.

Customizing the SpeechRecognition module

The SpeechRecognition module allows you to customize the speech recognition process by setting various parameters. For example, you can set the language of the speech, the threshold for recognizing speech, and the timeout for the recognition process.

Here’s an example of how to customize the Recognizer object:


import speech_recognition as sr

# Create a recognizer object
r = sr.Recognizer()

# Set the language to English
r.language = "en-US"

# Set the threshold for recognizing speech
r.energy_threshold = 300

# Set the timeout for the recognition process
r.operation_timeout = 10

In this example, the language is set to English (en-US), the threshold for recognizing speech is set to 300, and the timeout for the recognition process is set to 10 seconds.

Additional resources

For more information on the SpeechRecognition module and its capabilities, you can refer to the official documentation: https://pypi.org/project/SpeechRecognition/

Here are some additional examples of using the SpeechRecognition module:

Recognizing speech from a file

To recognize speech from a file, you can use the AudioFile class and the record() method:


import speech_recognition as sr

# Create a recognizer object
r = sr.Recognizer()

# Load the audio file
with sr.AudioFile("speech.wav") as source:
    audio = r.record(source)

# Recognize the speech
text = r.recognize_google(audio)

print(f"You said: {text}")

In this example, the AudioFile class is used to load the audio file speech.wav, and the record() method is used to record the audio from the file. Then, the recognize_google() method is used to recognize the speech.

Recognizing speech from a URL

To recognize speech from a URL, you can use the AudioFile class and the record() method, just like in the previous example. However, instead of loading a local file, you can use the URL class to download the audio file from a URL:


import speech_recognition as sr

# Create a recognizer object
r = sr.Recognizer()

# Download the audio file from a URL
url = "https://www.example.com/speech.wav"
audio = r.record(sr.AudioFile(url))

# Recognize the speech
text = r.recognize_google(audio)

print(f"You said: {text}")

In this example, the URL class is used to download the audio file from the URL https://www.example.com/speech.wav, and the record() method is used to record the audio from the file. Then, the recognize_google() method is used to recognize the speech.