Smart AI-based Chatbot using Intel MKL Library and NLP
- 0 Collaborators
A conversational Chatbot using Speech Recognition APIs and pre-trained Transformer models with and without intel toolkit. ...learn more
Project status: Under Development
oneAPI, Artificial Intelligence
Intel Technologies
MKL
Overview / Usage
The given code only contains the implementation of a basic chatbot that can perform some simple tasks like responding to greetings, telling the time, and engaging in a conversation using the DialoGPT model.
Methodology / Approach
My methodology involves building a conversational AI chatbot using Python and various frameworks and libraries such as SpeechRecognition, Pyttsx3, Transformers, and Intel MKL library.
To achieve this, I have used the DialoGPT-medium model from Microsoft and the transformers library for Python. The DialoGPT-medium model is a pre-trained neural network that has been fine-tuned to generate human-like responses to input text. The transformers library provides an easy-to-use interface for interacting with NLP models like DialoGPT-medium.
I have also incorporated speech recognition and text-to-speech conversion capabilities into the chatbot using the SpeechRecognition and pyttsx3 libraries. These libraries enable the chatbot to listen to and respond to spoken input from the user.
To handle different types of user requests, I have implemented various techniques such as keyword matching and regular expressions to identify specific keywords or phrases in the user's input. For example, if the user mentions the word "time," the chatbot will respond with the current time.
I have also included a polite response function that randomly selects a response from a list of predefined responses to express gratitude or acknowledge user input.
Intel MKL library is a highly optimized library for linear algebra and other mathematical operations. It provides optimized implementations of key linear algebra routines, which can significantly improve the performance of machine learning algorithms. In the code, I have used the MKL optimized random function, mkl_random.choice(), to select a polite response from a list of possible responses.
In summary, my methodology involves using a combination of Python libraries and frameworks to build a conversational AI chatbot, and leveraging the power of Intel MKL library to optimize the performance of the model. By using these tools and techniques, I am able to create a chatbot that can effectively understand and respond to human speech.