Speech-To-Text

Convertion of spoken words into text.

This project involves the following steps:

Step 1: Libraries

• speech_recognition
• AudioSegment from pydub
• requests
• os

Step 2: Function to download the audio file from a URL

• A URL as input
• An HTTP GET request to download the audio file
• Store it in memory using a BytesIO object
• Return the audio data, if download success. Otherwise, raise an exception.

Step 3: Function to convert the audio from MP3 to WAV

• Convert an MP3 file to WAV format
• Use pydub's AudioSegment to read the MP3 data
• Export it as a WAV file
• Store the WAV data in a BytesIO object.

Step 4: Function to transcribe audio file using CMU Sphinx

• Create a Recognizer object
• Read the audio file into memory
• Try to recognize the speech:
• If successful, return the transcribed text. If not, handle errors and return appropriate messages.

Step 5: Execution

• Specify the URL of the MP3 file to be downloaded.
• Download the audio file.
• If successful, convert the MP3 to WAV.
• If the conversion is successful, transcribe the audio
• Print the resulting text.