Scribd
Upload
5 views

Setting Up Packages for Speech Recognition

This document provides a step-by-step guide for setting up packages necessary for speech recognition, including activating the environment, installing audio processing libraries, and preparing for Whisper installation. It outlines specific commands for installing packages like librosa, SpeechRecognition, and Whisper AI, as well as additional setup for FFmpeg and PyTorch. Finally, it includes a verification step to ensure Whisper is correctly installed.

Uploaded by

rndattaba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views

Setting Up Packages for Speech Recognition

This document provides a step-by-step guide for setting up packages necessary for speech recognition, including activating the environment, installing audio processing libraries, and preparing for Whisper installation. It outlines specific commands for installing packages like librosa, SpeechRecognition, and Whisper AI, as well as additional setup for FFmpeg and PyTorch. Finally, it includes a verification step to ensure Whisper is correctly installed.

Uploaded by

rndattaba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Setting Up Packages for

Speech Recognition

Step 1: Ensure Your Environment is Active


• Open Anaconda Prompt (or Terminal if you’re using macOS).

• Confirm that you are in the correct environment. You should see the environment name
(e.g., (speech_env)) in parentheses.

• If you’re not in the correct environment, activate it with:

conda activate speech_env

Step 2: Install Audio Processing and Visualization


Packages
1. Install librosa (for audio processing):

o Run the following command in Anaconda Prompt (or Terminal if you’re using
macOS):

pip install librosa

2. Install SpeechRecognition (for converting spoken language to text):

o Enter the following command:

pip install SpeechRecognition

o Note: Be careful with the capitalization and spacing—SpeechRecognition must


be typed as a single word.

3. Install jiwer (for evaluating speech recognition accuracy):

o Type the following command:

pip install jiwer

4. Install matplotlib (for visualizations):

o Use this command to install:

pip install matplotlib

5. Install Google Text-to-Speech (gTTS) (for converting text into speech):

o Run this command:


pip install gTTS

Step 3: Prepare for Whisper Installation


Whisper by OpenAI requires some additional setup to handle multimedia data effectively.
Follow these steps to ensure everything is in place:

1. Install PyTorch:

o Go to the PyTorch website and select the “Get Started” option.

o Under “Start locally”, configure the following:

▪ Stable version

▪ Your operating system (e.g., Windows, macOS)

▪ Package: pip

▪ Language: Python

▪ Compute platform: Select a CUDA version if you have a GPU or select


CPU.

o Copy the provided installation command and paste it into Anaconda Prompt,
then press Enter.

2. Install Chocolatey (Windows only):

o Go to the Chocolatey website and click Install.

o Choose Individual installation and copy the provided code.

o Open Windows PowerShell as an administrator, paste the code, and run it.

Note: macOS users can use Homebrew instead of Chocolatey for the following steps.

3. Install FFmpeg:

o In Anaconda Prompt (or PowerShell with Chocolatey), type:

choco install ffmpeg

o Confirm with “Y” when prompted. FFmpeg is essential for Whisper to handle
various audio formats.

Step 4: Install Whisper AI for Speech Recognition

• With the environment activated in Anaconda Prompt, install Whisper using the
following command:

pip install -U openai-whisper


• Troubleshooting: If you receive an error stating that Whisper isn’t found after
installation, you can also try running the above pip install command in Command
Prompt instead of Anaconda Prompt.

Step 5: Verify Whisper Installation

• To confirm Whisper is correctly installed, type:

pip show openai-whisper

o If installed, this command will display Whisper’s details.

You’re all set! With all packages installed, you’re ready to dive into speech-to-text exercises.

You might also like