Created September 4, 2024

Seeing AI: Android App for Visually Impaired

Seeing AI is an Android app that uses OCR and TTS to convert text from images into audio.

Things used in this project

Hardware components

Nordic Semiconductor nRF52 Development Kit

Espressif ESP8266 ESP-01

Software apps and online services

Android Studio

Story

The Seeing AI project is an Android TTS (Text-to-Speech) OCR (Optical Character Recognition) Converter System designed to assist individuals with visual impairments by converting printed text into spoken words. This innovative application addresses the growing need for accessible reading solutions as the number of visually impaired individuals increases due to various factors.

System Functionality

The system operates by capturing images through the device's camera. It employs the Google Cloud Vision API for OCR, which accurately detects and extracts text from the captured images. Once the text is identified, it is converted into speech using TTS technology, allowing users to listen to the content through the device's speaker or headphones.

Key Features

User-Friendly Interface: The application is designed for simplicity, allowing navigation using the device's volume buttons. For instance:

Pressing the volume down once repeats the text.
Pressing it twice captures a new image.
Pressing the volume up exits the application.
User-Friendly Interface: The application is designed for simplicity, allowing navigation using the device's volume buttons. For instance:
Pressing the volume down once repeats the text.
Pressing it twice captures a new image.
Pressing the volume up exits the application.
Versatile Text Recognition: The system can read various types of text, including printed documents, digital displays, and signboards, providing on-the-go assistance.
Technology Stack: Developed using Java and Android Studio, the front end is created with XML, while the back end utilizes the Google Cloud Vision API for OCR and the built-in TTS capabilities of Android devices.

Advantages

Increased Independence: The system empowers visually impaired individuals to interact more freely with their environment by enabling them to read texts that would otherwise be inaccessible.
Portable and Convenient: As a mobile application, it can be used anywhere, making it a practical solution for daily challenges faced by visually impaired users.

Limitations

Despite its advantages, the system has some limitations:

Image Capture Quality: If the image is not captured properly, the OCR process may yield inaccurate results.
Volume Button Navigation: While intuitive for some, the volume button feature may be confusing for others, potentially hindering usability.

In conclusion, the Seeing AI project represents a significant advancement in assistive technology for visually impaired individuals. By transforming text into speech, it enhances access to information and fosters greater independence, ultimately improving the quality of life for users.

1 / 2

Seeing AI

public class SeeingAIActivity extends AppCompatActivity {

    private static final int REQUEST_IMAGE_CAPTURE = 1;
    private static final int REQUEST_PERMISSION = 2;

    private TextToSpeech textToSpeech;

    @Override
    protected void onCreate(Bundle savedInstanceState) {
        super.onCreate(savedInstanceState);
        setContentView(R.layout.activity_seeing_ai);

        // Initialize TextToSpeech
        textToSpeech = new TextToSpeech(this, status -> {
            if (status == TextToSpeech.SUCCESS) {
                // TTS initialization successful
            }
        });

        // Request camera permission
        if (ContextCompat.checkSelfPermission(this, Manifest.permission.CAMERA) != PackageManager.PERMISSION_GRANTED) {
            ActivityCompat.requestPermissions(this, new String[]{Manifest.permission.CAMERA}, REQUEST_PERMISSION);
        } else {
            captureImage();
        }
    }

    private void captureImage() {
        Intent takePictureIntent = new Intent(MediaStore.ACTION_IMAGE_CAPTURE);
        if (takePictureIntent.resolveActivity(getPackageManager()) != null) {
            startActivityForResult(takePictureIntent, REQUEST_IMAGE_CAPTURE);
        }
    }

    @Override
    protected void onActivityResult(int requestCode, int resultCode, @Nullable Intent data) {
        super.onActivityResult(requestCode, resultCode, data);
        if (requestCode == REQUEST_IMAGE_CAPTURE && resultCode == RESULT_OK) {
            Bundle extras = data.getExtras();
            Bitmap imageBitmap = (Bitmap) extras.get("data");

            // Perform OCR on the captured image
            String recognizedText = performOCR(imageBitmap);

            // Convert recognized text to speech
            textToSpeech.speak(recognizedText, TextToSpeech.QUEUE_FLUSH, null, null);
        }
    }

    private String performOCR(Bitmap bitmap) {
        // Use Google Cloud Vision API or other OCR libraries to extract text from the image
        // Return the recognized text as a string
        return "This is a sample recognized text.";
    }
}

Credits

Kirthana L

1 project • 1 follower

Embed the widget on your own site

Seeing AI: Android App for Visually Impaired

Seeing AI: Android App for Visually Impaired

Things used in this project

Hardware components

Software apps and online services

Story

System Functionality

Key Features

Advantages

Limitations

Code

Seeing AI

Credits

Kirthana L

Comments