Desktop Bot
| dc.contributor.author | MANISH PAUL1NH21CS152 SAHIL KUMAR AGRAWAL1NH21CS206 Aastha Bharadwaj1NH21IS001 Abhinav Kumar 1NH21IS003 | |
| dc.date.accessioned | 2025-04-23T06:23:07Z | |
| dc.date.available | 2025-04-23T06:23:07Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | The voice recognition system for the desktop bot is designed to make human-computer interaction smooth by using advanced machine learning techniques. It integrates the Google Speech-to-Text API with deep learning models for accurate and efficient speech recognition. Recurrent Neural Networks (RNNs) and their variants, such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs), process sequential speech data by capturing temporal dependencies. Complementing this, Convolutional Neural Networks (CNNs) extract features from audio spectrograms, identifying key patterns in frequency and amplitude. Attention mechanisms enhance the system's focus on critical sections of input, improving recognition precision. End-to-end models, including sequence-to-sequence (seq2seq) with attention and the Transformer model, map raw audio to text directly, streamlining the process and boosting efficiency. Specialized models turn raw sounds into meaningful components, while predictive models guess word sequences based on context, improving overall accuracy. This combination of techniques makes the desktop bot highly accurate and reliable, offering natural and intuitive voice-based human-computer interaction. | |
| dc.identifier.uri | http://192.168.75.5:4000/handle/123456789/17796 | |
| dc.language.iso | en | |
| dc.publisher | NHCE | |
| dc.title | Desktop Bot | |
| dc.type | Other |