Scribd
Upload
3 views2 pages

Object Classification & TTS Setup Guide

The document provides instructions for setting up object classification and text-to-speech functionality on a Raspberry Pi 4. It includes steps for enabling a virtual environment, libraries used, camera setup, and text-to-speech engine initialization. Additionally, troubleshooting tips are offered for common issues related to directory navigation, camera indexing, and virtual environment activation.

Uploaded by

venomiscool71
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3 views2 pages

Object Classification & TTS Setup Guide

The document provides instructions for setting up object classification and text-to-speech functionality on a Raspberry Pi 4. It includes steps for enabling a virtual environment, libraries used, camera setup, and text-to-speech engine initialization. Additionally, troubleshooting tips are offered for common issues related to directory navigation, camera indexing, and virtual environment activation.

Uploaded by

venomiscool71
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Documentation: Object Classification and Text to Speech Functionality.

First go to directory where your project resides:


Command: “cd ~/projects/tflite1”

1. Enabling virtual environment within raspberry 4


Command:
workon pyenv
Purpose:
This command enables virtual environment in python
2. Libraries used
a. Pyttsx3: This is a text to speech engine library that enables text to speech
functionality
b. cv2(opencv): Used for capturing video, resizing frames and processing frames.
c. Numpy: used for efficient handling of multidimensional data structures like array.
d. Tensorflow: open source library used for diving into machine learning.

3. Setting up the camera.


def __init__(self,resolution=(640,480),framerate=30):
# Initialize the PiCamera and the camera image stream
[Link] = [Link](0)
ret = [Link](cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*'MJPG'))
ret = [Link](3,resolution[0])
ret = [Link](4,resolution[1])

# Read first frame from the stream


([Link], [Link]) = [Link]()

# Variable to control when the camera is stopped


[Link] = False

“ [Link] = [Link](0)” - Make sure to put the corresponding camera


index. Ranges between 0 - 5.

4. Text to Speech Engine Initialization


def init_engine():
engine = [Link]()
[Link]('rate', 100)
[Link]('volume',1.0)
return engine
Command to run the code:
python3 TFLite_detection_webcam.py --
modeldir=Sample_TFLite_model

Note: modeldir is the directory where your local model is stored. In your
raspi, There is directory Sample_TFLite_model directory

Troubleshooting.
1. Make sure you are in the projects directory before interacting.
2. Camera error: Make sure to place the right camera index. Refer to setting up camera
section.
3. Environment error: Error indicating module not found or library not found often refers to
the problem of not activating virtual environment. Make have activated before running
the script

You might also like