The classification of sentiment on a small scale often suffers from the small amount of data, which limits the generalization ability of the models. This study evaluates and compares the effectiveness of three data augmentation strategies:

• EDA (Easy Data Augmentation)
• Back-translation
• Contextual token substitution (nlpaug-style)

Methods: Tested on both traditional ML classifiers (SVM, Random Forest) and transformer-based models (BERT) using low-resource sentiment datasets.

• All augmentation methods improved performance
• Contextual augmentation gave the most consistent gains for BERT
• EDA and back-translation were more effective for traditional classifiers

Lip reading plays a crucial role in applications such as speech recognition, assistive technologies for the hearing-impaired, and biometric authentication. However, performance often degrades when the speaker varies, the environment changes, or the speech sounds are visually similar. To overcome these challenges, we propose LipLingo:

• Based on the LipNet architecture
• Enhanced with standardized preprocessing for consistent mouth-region representation
• Uses rotational validation for better generalization
• Combines 3D convolutional layers (spatio-temporal features) and bidirectional recurrent layers (sequence modeling)
• Optimized with Adam optimizer and CTC loss

Results:

• 95% character-level accuracy
• 87% sentence-level accuracy
• Outperforms baseline LipNet