project name : Voice assistant powered by Pytorch
Team members: 1)Harini I (Sri sairam engineering college,India) 2)Balakrishnan V (St.Joseph's College of engineering,India)
We have done a simple voice assistant application which basically handles customer queries.Basically our concept is to make this as a shop assistant but we also added some cool stuffs so that it works as a basic voice assistant . We used Natural Language Tool kit which helps us to understand the text receive valuable insights. NLP tools give us a better understanding of how the language may work in specific situations. Moreover, people also use it for different business purposes. Such proposes might include data analytics, user interface optimization, and value proposition. But, it was not always this way.
Natural Language Toolkit features include:
Text classification Part-of-speech tagging Entity extraction Tokenization Parsing Stemming Semantic reasoning
We used pretrained model .
Processes involved: 1)Tokenization 2)stemming 3)creating bag of words (array of numbers) 4)creating intents Eg: tag:'greeting'->pattern:'hi how are you','Hey hi'->response:'Hey hi','hello'. 5)creating training data 6)load the model.
we used speech recognition module to get input from the user and done the above mentioned processes.Appropriate response from the intent file is chosen by our model and the text format of the output is converted to speech.
Features: 1)Sending email 2)Opening browsers 3)playing songs 4)Getting feedback from the user when customer leave the shop.
We also done the front end of our application but there is a problem in our flask sever.We currently working on it to fix.
We also trained YOLO v5 to classify the gender so that it can address customers which will be so good.But we couldn't link that with our project .we are beginners so we are suffering a lot to do this kind of things like creating API of our trained model and using it in our another project. We hope we have done the best which we can do. Really learned a lot of new things.And had a great time in participating in pytorch summer hackathon 2020.