Publications

LoRA Fine-Tuning

Fine-tuned RoBERTa using LoRA on the MultiNLI dataset.

Developed a Persian voice chatbot integrated with Whisper for speech recognition and ChatGPT for conversational responses.

Fine-tuned Whisper model for Persian language.

Fine-tuned the HuBERT model on the SheMO: Persian Speech Emotion Detection Database.

Fine-tuned SAM on a custom dataset for detecting bodies of water.

An emotionally aware robot assistant capable of receiving voice commands

Implemented the MUSIC algorithm to estimate the Direction of Arrival (DOA) of sound sources

Developed a custom Faster R-CNN implementation for wildfire smoke detection.

Developing a model to classify Persian music Dastgah.