دانلود مقاله مطالعه تشخیص احساسات موسیقی
چکیده
مقدمه
دانش مقدماتی
استخراج ویژگی و شناسایی احساسات
وضعیت و روند توسعه
نتیجه گیری
منابع
Abstract
Introduction
Preliminary knowledge
Feature extraction and emotion recognition
Development status and trends
Conclusions
References
چکیده
موسیقی زبان احساسات است. در سال‌های اخیر، تشخیص احساسات موسیقی توجه گسترده‌ای را در جامعه دانشگاهی و صنعتی به خود جلب کرده است، زیرا می‌تواند به طور گسترده در زمینه‌هایی مانند سیستم‌های توصیه، آهنگسازی خودکار موسیقی، روان‌درمانی، تجسم موسیقی و غیره استفاده شود. به خصوص با توسعه سریع هوش مصنوعی، تشخیص احساسات موسیقی مبتنی بر یادگیری عمیق به تدریج در حال تبدیل شدن به جریان اصلی است. این مقاله بررسی دقیقی از تشخیص احساسات موسیقی ارائه می دهد. این مقاله با شروع برخی از دانش اولیه در مورد تشخیص احساسات موسیقی، ابتدا برخی از معیارهای ارزیابی رایج را معرفی می کند. سپس یک چارچوب تحقیقاتی سه بخشی ارائه می شود. بر اساس این چارچوب تحقیقاتی سه بخشی، دانش و الگوریتم های درگیر در هر بخش با تجزیه و تحلیل دقیق، از جمله برخی از مجموعه داده های رایج، مدل های احساسی، استخراج ویژگی و الگوریتم های تشخیص احساسات، معرفی می شوند. پس از آن، مشکلات چالش برانگیز و روند توسعه فناوری تشخیص احساسات موسیقی پیشنهاد می شود و در نهایت، کل مقاله خلاصه می شود.
توجه! این متن ترجمه ماشینی بوده و توسط مترجمین ای ترجمه، ترجمه نشده است.
Abstract
Music is the language of emotions. In recent years, music emotion recognition has attracted widespread attention in the academic and industrial community since it can be widely used in fields like recommendation systems, automatic music composing, psychotherapy, music visualization, and so on. Especially with the rapid development of artificial intelligence, deep learning-based music emotion recognition is gradually becoming mainstream. This paper gives a detailed survey of music emotion recognition. Starting with some preliminary knowledge of music emotion recognition, this paper first introduces some commonly used evaluation metrics. Then a three-part research framework is put forward. Based on this three-part research framework, the knowledge and algorithms involved in each part are introduced with detailed analysis, including some commonly used datasets, emotion models, feature extraction, and emotion recognition algorithms. After that, the challenging problems and development trends of music emotion recognition technology are proposed, and finally, the whole paper is summarized.
Introduction
In recent years, the electronic music market has achieved rapid development, massive music resources can be obtained from various sources. These music resources need to be organized and managed based on label information such as emotion, genre, etc. so that listeners can obtain music works conveniently. Since music is the carrier of emotions, so it is particularly important to recognize the emotion labels of music works. Using manual methods to obtain the label information can be time-consuming, labor-intensive, and error-prone. Therefore, the research field of automatically recognizing emotion labels came into being.
Conclusions
This paper reviews the current research on MER. Firstly, it introduces the research background, gives the definition, summarizes the significance of MER, and gives a brief introduction of MER history. Then the current research framework is introduced, and the knowledge and algorithms involved in each part are elaborated. Lastly, the challenging problems and future development trends of MER are pointed out.
There are two main contributions of this paper. The first is to give a detailed analysis of research papers that use the DL technique, the uniqueness, models, and experiments of each paper are elaborated. Secondly, the challenging problems faced by MER and future development trends are pointed out in Section 4. Currently, in the field of MER, there are urgent needs for authoritative large-scale diversified datasets and more accurate emotion models. Music concepts and carefully designed features are also needed. Generally speaking, the MER field is shifting from static processing to dynamic process, from single modal to multi-modal and from traditional ML models to DL models, more technologies such as transfer learning and knowledge graphs, more information like the singing voice, social tags, album cover data, and MV data can be explored, and ideas from related fields including AME, CR, and MGC can be borrowed.