International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
Human Visuals for Vibrant Keytunes – HVVK an Emotion Detection Music Recommendation System using Spotify API Vania Panjwani1, Harsha Gotmare2, Vedant Masane3 , Prof. Dr Rashmi Jaiswal4 1Student,Dept.of Comp Sci & engineering, COETA, Akola, Maharashtra, India
2 Student, Dept. of Comp Sci & engineering, COETA, Akola, Maharashtra, India 3 Student, Dept. of Comp Sci &engineering, COETA, Akola, Maharashtra, India
4 Assistant Professor, Dept. of Comp Sci & engineering, COETA, Akola, Maharashtra, India ---------------------------------------------------------------------***--------------------------------------------------------------------experience listening to music. As everyone knows, our Abstract - The swift progress in mobile and internet
feelings have a big impact on the kind of music we listen to. For example, someone who is happy could enjoy lively, uplifting music, but someone who is depressed might prefer more reflective or peaceful sounds. Conventional music recommendation systems could miss the user's current emotional state because they frequently rely on the user's past and preferences. Our approach uses AI, Spotify's vast music catalog, and facial emotion recognition to overcome this constraint.
technology has granted us unrestricted access to an extensive array of music resources. In the music industry, certain musical genres could be more well-liked than others. At the moment, users of music-listening apps have to make special updates to the static playlists to reflect their own preferences. Music recommendation systems have become an integral part of music listening. However, most traditional recommendation systems rely on user activity data or metadata, which is not enough to capture the emotional resonance of music. Given that music has a strong emotional impact on listeners, personalized music recommendation systems that include users' emotional states are highly desired.
2. PROBLEM STATEMENT Even though music evolved a lot, streaming services have also seen new developments, but still users face the challenge of not discovering music that matches their mode and emotion. Traditional music recommendation was based on physical input from the user or from the previous metadata that provided recommendations to the player and then to the user. Music has a strong emotional impact on the listeners, but the traditional system fails to capture the emotions of the listener. Therefore, the aim of this paper is to create a system that uses machine learning algorithms to present a cross-platform music player that makes recommendations for music depending on the user's current mood as seen through a webcam.
Two different models and methods were available for this purpose: FER (Facial Emotion Recognition), which detects mood from facial expressions, and Music Classification Models, which select songs. To provide the user with recommendations that are accurate and efficient, we are trying to integrate these two systems. To recommend music that is appropriate for the user's emotions, we will be developing a system in this project that will allow us to collect the user's real-time emotions through conversation or other methods. Key Words: Recommendation System, Facial Emotion Recognition, Interactive UI, Mood-based music classifier, Spotify API.
3. LITERATURE REVIEW Facial expressions convey the person's present mental state. When expressing feelings to others, we typically do it using nonverbal cues including tone of voice, facial expressions, and hand gestures. Preema et al. [1] claimed that making and maintaining a big playlist takes a lot of effort and time. According to the publication, the music player chooses a tune based on the user's present mood. Playlists based on mood are created by the application by scanning and categorizing audio files based on audio attributes. The Viola-Jonas method, which is used for face detection and facial emotion extraction, is utilized by the program. The categorization process employed Support Vector Machine (SVM) to extract data into five main universal emotions, such as anger, joy, surprise, sad, and disgust.
1. INTRODUCTION Music now plays a vital role in our lives in the digital age by providing a soundtrack for our feelings and experiences. With the introduction of streaming services, we now have an incredible amount of music at our fingertips. But finding music that speaks to our current emotional states and connecting with them is the difficult part—not the availability of music. In an attempt to create customized playlists that suit a listener's tastes and mood, music recommendation algorithms have surfaced as a solution to this problem. The main goal of this project is to close the gap that exists between a user's emotional state and their
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 2446