In today's world where everyone is striving for achievement, but many people are succumbing to depression as a result of this struggle. Making a few poor decisions occasionally. As a result, Moodify was designed to assist people in de-stressing by recommending songs based on their feelings.

Moodify is a facial expression recognition-based music suggestion website that cheer up users and saves time while searching for a song that matches their mood.

1. It recognizes facial expression based on the 7 categories i.e., angry, sad, fear, happy, disgust, surprise and neutral.
2. Based on the emotion it gives user two choices either suggesting songs.
3. If user wishes to watch songs then a list of songs matching their mood are suggested with songs poster.
4. When user clicks on songs it redirects them to Spotify website.

Python is the programming language used to create the emotion recognition model and deploy it on the web application using flask. CV2, TensorFlow, NumPy, matplotlib, and other libraries are also utilized. The model is build using the transfer learning approach for which MobileNet model is used. The FER-2013 dataset, which comprises around 35000 photos, was utilized for model training and validation. This model is deployed on a website created with HTML and CSS using the flask framework. Based on the seven emotions, a new dataset of music was constructed. The data from songs was utilized to create the various templates that correlate to various emotions.

It was hard to find Songs datasets that reflected different emotions on the web. Finding templates of emotion-based songs proved to be a major challenge. It was a challenging task to get the website to access 7 songs templates from the song's button based on the emotion recognized. Dynamic links were used as a means to access the templates for songs corresponding to the output of the model.

Creating a project that is effective for the modern generation. Working in unison despite being far away from each other. Creating this awesome project in such a small span of time! Learnt a great deal about Python and different frameworks such as flask and its integration, many new libraries in Python. We thought and build the entire solution in such a small amount of time, overcame all the challenges and find a hack to each problem.

We learnt about the transfer learning approach for the model, which assisted us in face identification, as well as how to develop basic web pages using HTML, CSS and how to deploy the model on the web using flask.

The next step is to improve the model's accuracy. We are also going to introduce an AI-based chat bot with which users may express their problems and, based on the same, the Bot will recommend some remedies as well as the most suitable song in that case.