TWI482149B - The Method of Emotional Classification of Game Music - Google Patents
The Method of Emotional Classification of Game Music Download PDFInfo
- Publication number
- TWI482149B TWI482149B TW101148696A TW101148696A TWI482149B TW I482149 B TWI482149 B TW I482149B TW 101148696 A TW101148696 A TW 101148696A TW 101148696 A TW101148696 A TW 101148696A TW I482149 B TWI482149 B TW I482149B
- Authority
- TW
- Taiwan
- Prior art keywords
- music
- game
- game music
- emotional
- parameters
- Prior art date
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Description
本發明是有關於一種音樂分類方法,特別是指一種遊戲音樂的情緒分類方法。The present invention relates to a music classification method, and more particularly to a method for classifying emotions of game music.
在資訊爆炸的時代中,音樂的獲取相當容易,配合電玩遊戲的日益興盛,許多電玩遊戲後來都會針對其使用之遊戲音樂另外發行遊戲原聲帶,或甚至舉辦演奏會。而遊戲音樂為了符合遊戲場景內容,通常都是有故事性,因此若能以情緒來分類遊戲音樂,並讓遊戲者可挑選符合自己心情的遊戲音樂,則該電玩遊戲將能更受遊戲者所喜愛。In the era of information explosion, the acquisition of music is quite easy. With the growing popularity of video games, many video games will later release game soundtracks for the game music they use, or even hold concerts. In order to match the content of the game scene, the game music is usually story-like. Therefore, if the game music can be classified by emotion and the player can select the game music that suits his mood, the game game will be more affected by the player. favorite.
雖然目前已有業者開發出所謂能夠對音樂進行情緒分類的軟體系統或裝置,但這類軟體系統或裝置的分類方法都是以預先設定之制式規則進行音樂情緒分類,由於每個人對於音樂的感受都不盡相同,其分類結果經常不被使用者所接受,所以這種制式分類方式並無法適用每個人,適用性差。Although there are currently developers who have developed so-called software systems or devices that can classify music emotions, the classification methods of such software systems or devices are based on pre-set rules of music to classify music emotions, because everyone feels about music. They are not the same, and their classification results are often not accepted by users. Therefore, this method of classification cannot be applied to everyone and its applicability is poor.
因此,本發明之目的,即在提供一種透過調性、音程與節奏對遊戲音樂進行情緒分類的情緒分類方法。Accordingly, it is an object of the present invention to provide an emotional classification method for emotionally classifying game music through tonality, interval and rhythm.
本發明之另一目的在於提供一種可供使用者根據自身感受進行遊戲音樂之情緒分類的情緒分類方法。Another object of the present invention is to provide an emotion classification method for a user to classify emotions of game music according to his or her own feelings.
於是,本發明遊戲音樂的情緒分類方法,適用於透過軟體及/或硬體的方式建構在一電子裝置中,而用以對遊戲 音樂進行情緒分類,包含以下步驟:(A)對該電子裝置儲存之多首遊戲音樂樣本分別設定一對應的情緒形容詞參數;(B)對步驟(A)之每一首遊戲音樂樣本進行調性、節奏與音程等音樂特徵分析,而對應每一首遊戲音樂樣本輸出多個音樂特徵參數;(C)以多類別支援向量機統計分析該等遊戲音樂樣本之情緒形容詞參數分別與對應之該等音樂特徵參數的關連性,而建構輸出一個多類別情緒預測模型;及(D)對該電子裝置後續儲存之每一首新遊戲音樂進行音樂特徵分析而對應輸出多個音樂特徵參數,並以該多類別情緒預測模型分析該新遊戲音樂之該等音樂特徵參數,而對該音遊戲音樂進行情緒分類,並輸出一對應該新遊戲音樂之情緒形容詞參數。Therefore, the emotional classification method of the game music of the present invention is applicable to being constructed in an electronic device through software and/or hardware, and is used for the game. Music classifying emotions includes the following steps: (A) respectively setting a corresponding emotional adjective parameter for the plurality of game music samples stored in the electronic device; (B) performing tonality on each of the game music samples of step (A) Music characteristics such as rhythm and interval, and outputting a plurality of music feature parameters for each game music sample; (C) statistically analyzing the emotional adjective parameters of the game music samples by the multi-category support vector machine respectively and corresponding thereto Correlation of music feature parameters, and constructing a multi-category emotion prediction model; and (D) performing music feature analysis on each new game music stored subsequent to the electronic device, and correspondingly outputting a plurality of music feature parameters, and The multi-category emotion prediction model analyzes the music feature parameters of the new game music, and emotionally classifies the music game music, and outputs a pair of emotional adjective parameters that should be new game music.
本發明之功效:藉由以支援向量機對遊戲音樂之調性、音程與節奏等音樂特徵參數,以及對該等遊戲音樂所設定之情緒形容詞參數進行統計分析的方式,可建構出一能夠對遊戲音樂進行情緒分類,且分類準確度高的多類別情緒預測模型。The effect of the invention: by using the support vector machine to analyze the musical characteristic parameters such as tonality, interval and rhythm of the game music, and the statistical analysis of the emotional adjective parameters set by the game music, Game music is a multi-category emotion prediction model that classifies emotions and has high classification accuracy.
有關本發明之前述及其他技術內容、特點與功效,在以下配合參考圖式之一個較佳實施例的詳細說明中,將可清楚的呈現。The above and other technical contents, features and advantages of the present invention will be apparent from the following detailed description of the preferred embodiments.
如圖1所示,本發明遊戲音樂的情緒分類方法的較佳實施例,適用於透過軟體及/或硬體的方式建構在一電子裝置中,而可用以對該電子裝置所儲存之遊戲音樂進行情緒 分類。該情緒分類方法包含以下步驟:步驟(一)對遊戲音樂樣本設定情緒形容詞參數。先對多首已經被確定其情緒分類之遊戲音樂樣本分別設定一代表其情緒歸屬的情緒形容詞參數。As shown in FIG. 1 , a preferred embodiment of the method for classifying emotions of the game music of the present invention is applicable to software and/or hardware built in an electronic device, and the game music stored in the electronic device can be used. Emotion classification. The emotion classification method comprises the following steps: Step (1) setting an emotional adjective parameter to a game music sample. First, a plurality of emotional adjective parameters representing the emotional attribution of the game music samples whose emotion classification has been determined are respectively set.
在本實施例中,會將該等遊戲音樂樣本分為五類情緒,且每一種情緒分類會對應一情緒形容詞與一情緒形容詞參數,該等情緒形容詞分別為『平靜的』、『冰冷的』、『奇異的』、『幽默的』,及『熱情的』。但實施時,採用之情緒形容詞的類型不以上述類型為限。In this embodiment, the game music samples are divided into five types of emotions, and each of the emotion categories corresponds to an emotional adjective and an emotional adjective parameter, and the emotional adjectives are “quiet” and “cold” respectively. "Singular", "humorous", and "enthusiasm". However, the type of emotional adjectives used is not limited to the above types.
步驟(二)對所有遊戲音樂樣本進行音樂特徵分析,以取得該等遊戲音樂樣本的音樂特徵參數。對每一首遊戲音樂樣本進行調性、音程與節奏等三種音樂特徵分析,而對應輸出每一首遊戲音樂樣本的音樂特徵參數。Step (2) Performing music feature analysis on all game music samples to obtain music feature parameters of the game music samples. Each of the game music samples is analyzed for three musical characteristics such as tonality, interval and rhythm, and the music characteristic parameters of each game music sample are output correspondingly.
其中,調性音樂特徵分析,是分析每一首遊戲音樂樣本中所含有之調性種類,及各種調性的分佔比例(%),例如C大調、D大調、E大調與B大調等,及C小調、D小調與B小調等調性的分佔比例,而統計分析出每一情緒形容詞參數所對應之所有遊戲音樂樣本的各種調性的平均分佔比例(%)。音程音樂特徵是分析每一首遊戲音樂樣本所採用之完全協和音程、不完全協和音程,及不協和音程的分佔比例,並統計分析出每一情緒形容詞參數所對應之所有遊戲音樂樣本的完全協和音程、不完全協和音程,及不協和音程之平均分佔比例(%)。該節奏音樂特徵的分析是分析每一首遊戲音樂樣本的節奏(beat per minute,bpm),並統計分 析出每一情緒形容詞參數所對應之所有遊戲音樂樣本的平均節奏速度。Among them, the analysis of tonal music characteristics is to analyze the tonal types contained in each game music sample, and the proportion of various tonality (%), such as C major, D major, E major and B Major adjustments, and the proportion of tonality in C minor, D minor and B minor, and statistical analysis of the average proportion (%) of various tonality of all game music samples corresponding to each emotional adjective parameter. The interval music feature is to analyze the complete consonant interval, incomplete consonant interval, and the proportion of non-coordinated intervals used in each game music sample, and statistically analyze the complete concord of all game music samples corresponding to each emotional adjective parameter. The interval (%) of the interval, the incomplete harmony interval, and the average proportion of the uncoordinated intervals. The analysis of the rhythm music feature is to analyze the beat per minute (bpm) of each game music sample, and statistical scores. The average rhythm velocity of all game music samples corresponding to each emotional adjective parameter is precipitated.
在分析出每一首遊戲音樂樣本之該等音樂特徵後,會將該等音樂特徵進行數據編碼與正規化處理,以產生所需的音樂特徵參數。After analyzing the music features of each game music sample, the music features are data encoded and normalized to produce desired music feature parameters.
在本實施例中,該等遊戲音樂樣本之調性音樂特徵、音程音樂特徵與節奏音樂特徵是分別透過MATLAB程式軟體進行分析,但實施時,由於前述音樂特徵之分析方法眾多,因此不以本案採用之軟體為限。此外,由於數據編碼與正規化為現有統計分析常用之技術手段,且方式眾多,因此不再詳述。In this embodiment, the tonal music features, the interval music features, and the rhythm music features of the game music samples are separately analyzed by the MATLAB software, but when implemented, due to the numerous analysis methods of the aforementioned music features, the case is not The software used is limited. In addition, since data coding and normalization are common technical means for existing statistical analysis, and there are many ways, they will not be described in detail.
步驟(三)以支援向量機(Support Vector Machine,SVM)建立多類別情緒預測模型。以支援向量機對步驟(一)每一遊戲音樂樣本所設定的情緒形容詞參數,及步驟(二)對該等遊戲音樂樣本分析所得之音樂特徵參數進行多類別情緒預測模型的建立。Step (3) establish a multi-category emotion prediction model with Support Vector Machine (SVM). The multi-category emotion prediction model is established by the support vector machine for the emotional adjective parameters set in step (1) for each game music sample, and (2) the music feature parameters obtained from the analysis of the game music samples.
在本實施例中,根據每一遊戲音樂樣本的該等音樂特徵參數,及每一遊戲音樂樣本對應的情緒形容詞參數進行統計分析,而形成多類別分類問題,再將多類別分類問題分解為一系列一對一(One-against-one,OAO)的支援向量機模型,並透過高斯核心函數(Gaussian kernel)進行交叉驗證,並根據所有支援向量機模型的判斷結果,而建立出一個多類別情緒預測模型。此時,該多類別情緒預測模型即可用以對該電子裝置另外儲存之其他遊戲音樂進行情緒 分類。In this embodiment, statistical analysis is performed according to the music feature parameters of each game music sample and the emotional adjective parameters corresponding to each game music sample, thereby forming a multi-category classification problem, and then decomposing the multi-category classification problem into one A one-to-one (One-against-one, OAO) support vector machine model, cross-validated by Gaussian kernel function, and based on the judgment results of all support vector machine models, a multi-category emotion is established. Forecast model. At this time, the multi-category emotion prediction model can be used to emotions other game music stored separately for the electronic device. classification.
步驟(四)分析該等音樂特徵參數對於每一情緒形容詞參數的重要性。根據步驟(三)所建立之多類別情緒預測模型,進一步透過支援向量機遞迴特徵消去法(support vector machine recursive feature elimination,SVMREF)分析每一個音樂特徵參數對每一情緒形容詞參數的影響權重,而輸出該等音樂特徵參數在每一情緒形容詞參數中的權重排列順序資料,例如表1所示。Step (4) analyzes the importance of the musical feature parameters for each emotional adjective parameter. According to the multi-category emotion prediction model established in step (3), the support vector machine recursive feature elimination (SVMREF) is further used to analyze the weight of each music feature parameter on each emotional adjective parameter. And outputting the weighting order data of the music feature parameters in each of the emotional adjective parameters, as shown in Table 1.
此權重排序順序可方便遊戲音樂製作者了解哪種音樂特徵對於哪一種情緒的表達較具有影響力,而有助於使製作出來的遊戲音樂所表達的情緒更符合遊戲劇情,也更容易引發遊戲者的情感共鳴。This weight sorting order can facilitate game music producers to understand which music features are more influential for the expression of which emotions, and help to make the emotions expressed by the produced game music more in line with the game story, and more likely to trigger the game. The emotional resonance of the person.
步驟(五)以該多類別情緒預測模型對該電子裝置儲存之其他遊戲音樂進行情緒分類。在根據前述遊戲音樂樣本建立該多類別情緒預測模型後,就可以該多類別情緒預測模型進行遊戲音樂的情緒分類處理。進行分類時,同樣是先對該等帶分類之新遊戲音樂進行音樂特徵分析,以取 得每一待分類之新遊戲音樂的調性、音程與節奏等音樂特徵參數,然後再以該多類別情緒預測模型對每一待分類之新遊戲音樂的該等音樂特徵參數進行分析,而找出與每一待分類之新遊戲音樂對應的情緒形容詞參數,進而分析出每一新遊戲音樂所屬的情緒類別,並於完成情緒分析後,驅使該電子裝置回饋顯示出一對應於該被分析之新遊戲音樂的情緒形容詞參數的動畫影像,藉以提醒該電子裝置使用者。Step (5) emotionally classifying the other game music stored in the electronic device by the multi-category emotion prediction model. After the multi-category emotion prediction model is established according to the foregoing game music sample, the multi-category emotion prediction model can perform the emotion classification processing of the game music. When classifying, the music features of the new game music with the classification are also analyzed first. The music feature parameters such as tonality, interval and rhythm of each new game music to be classified are analyzed, and then the music feature parameters of each new game music to be classified are analyzed by the multi-category emotion prediction model, and Equivalent adjective parameters corresponding to each new game music to be classified, and then analyzing the emotion categories to which each new game music belongs, and after completing the emotion analysis, driving the electronic device to feedback to display a corresponding to the analyzed An animated image of the emotional adjective parameters of the new game music to alert the user of the electronic device.
步驟(六)將分析後之新遊戲音樂依據其情緒形容詞參數進行分類儲存,建立一遊戲音樂情緒分類資料庫。Step (6) classifying and storing the new game music after analysis according to the parameters of the emotional adjectives, and establishing a game music emotion classification database.
本發明透過上述步驟對該等遊戲音樂樣本進行調性、音程與節奏等音樂特徵參數分析,並配合五種情緒形容詞參數進行支援向量機統計分析,以建構出多類別情緒預測模型,此多類別情緒預測模型之情緒分類準確度高,確實可有效根據遊戲音樂的該等音樂特徵參數進行準確的情緒分類。Through the above steps, the present invention analyzes the musical feature parameters such as tonality, interval and rhythm of the game music samples, and performs statistical analysis of the support vector machine with five kinds of emotional adjective parameters to construct a multi-category emotion prediction model. The emotion prediction model has high emotional classification accuracy, and it can effectively perform accurate emotion classification according to the music characteristic parameters of the game music.
必須說明的是,前述遊戲音樂樣本之初步情緒分類,可以是使用建構有本方法之電子裝置之使用者依據其本身聆聽感受所做之分類,所以步驟(三)所建立的多類別情緒預測模型的遊戲音樂情緒分類模式會貼近該位使用者之個人喜好,因此,本發明遊戲音樂的情緒分類方法可方便建構有本方法之電子裝置的使用者,依據其自身感受來建立專屬的多類別情緒預測模型,使遊戲音樂的分類結果更符合使用者的需求。It should be noted that the preliminary emotional classification of the aforementioned game music sample may be a classification made by the user who constructs the electronic device with the method according to his own listening feeling, so the multi-category emotion prediction model established in step (3) The game music emotion classification mode will be close to the personal preference of the user. Therefore, the emotion classification method of the game music of the present invention can conveniently construct the user of the electronic device of the method, and establish an exclusive multi-category emotion according to the feeling of the user. The prediction model makes the classification result of the game music more in line with the needs of the user.
當然,實施時,該等遊戲音樂樣本的情緒分類也可以是透過問卷實驗分析所產生,此時,透過本方法所建構之多類別情緒預測模型的遊戲分類結果,則會較貼近一般大眾的普遍性認知,而可適用於遊戲軟體開發者使用。Of course, during the implementation, the emotional classification of the game music samples can also be generated through the analysis of the questionnaire. At this time, the game classification results of the multi-category emotion prediction model constructed by the method will be closer to the general public. Sexual cognition, which can be applied to game software developers.
因此,本發明遊戲音樂的情緒分類方法可方便根據使用者初始建立之遊戲音樂的情緒分類差異,來建立出符合各種需求之專屬多類別情緒預測模型,而可改善傳統音樂情緒分類系統僅能依據其原先設定之制式分析模式進行音樂情緒分類的缺點。Therefore, the emotion classification method of the game music of the present invention can conveniently establish an exclusive multi-category emotion prediction model that meets various needs according to the emotional classification difference of the game music initially established by the user, and can improve the traditional music emotion classification system only according to the basis. The previously set system analysis mode has the disadvantage of classifying music emotions.
綜上所述,藉由以支援向量機對遊戲音樂之調性、音程與節奏等音樂特徵參數,以及對該等遊戲音樂所設定之情緒形容詞參數進行統計分析的方式,可建構出一能夠對遊戲音樂進行情緒分類,且分類準確度高的多類別情緒預測模型,此將有助於電玩遊戲者與電玩遊戲開發者快速獲取遊戲音樂之情緒分類,並瞭解各種音樂特徵參數對於情緒屬性的影響力,遊戲音樂開發者可針對特定遊戲情感內容或情境,設計出符合玩家需求的遊戲產品。此外,本發明也可供使用者依據其自身聆聽感受所做之遊戲音樂樣本的情緒分類,透過本發明建立出一個專屬的多類別情緒預測模型,而可使日後對遊戲音樂的情緒分類結果更貼近本身喜好,而可改善傳統音樂情緒分類系統僅能依據其原先設定之制式分析模式進行音樂情緒分類的缺點,相當方便實用。因此,確實可達到本發明之目的。In summary, by using the support vector machine to analyze the musical characteristic parameters such as tonality, interval and rhythm of the game music, and the statistical analysis of the emotional adjective parameters set by the game music, it is possible to construct a Game music is a multi-category emotion prediction model with emotional classification and high classification accuracy, which will help video game players and video game developers to quickly obtain the emotional classification of game music and understand the influence of various music feature parameters on emotional attributes. Force, game music developers can design game products that meet the needs of players for specific game emotional content or situations. In addition, the present invention can also be used by the user to establish an exclusive multi-category emotion prediction model according to the emotional classification of the game music samples made by the user according to his own listening experience, so that the emotional classification result of the game music can be further improved in the future. Close to their own preferences, but can improve the traditional music sentiment classification system can only be based on its original set of analysis mode for music emotional classification of the shortcomings, quite convenient and practical. Therefore, the object of the present invention can be achieved.
惟以上所述者,僅為本發明之較佳實施例而已,當不 能以此限定本發明實施之範圍,即大凡依本發明申請專利範圍及發明說明內容所作之簡單的等效變化與修飾,皆仍屬本發明專利涵蓋之範圍內。However, the above is only the preferred embodiment of the present invention, when not The scope of the invention is to be construed as being limited by the scope of the invention and the scope of the invention.
圖1是本發明遊戲音樂的情緒分類方法之一較佳實施例的步驟流程圖。BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a flow chart showing the steps of a preferred embodiment of the emotion classification method for game music of the present invention.
Claims (3)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW101148696A TWI482149B (en) | 2012-12-20 | 2012-12-20 | The Method of Emotional Classification of Game Music |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW101148696A TWI482149B (en) | 2012-12-20 | 2012-12-20 | The Method of Emotional Classification of Game Music |
Publications (2)
Publication Number | Publication Date |
---|---|
TW201426731A TW201426731A (en) | 2014-07-01 |
TWI482149B true TWI482149B (en) | 2015-04-21 |
Family
ID=51725633
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW101148696A TWI482149B (en) | 2012-12-20 | 2012-12-20 | The Method of Emotional Classification of Game Music |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI482149B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7223848B2 (en) * | 2018-11-15 | 2023-02-16 | ソニー・インタラクティブエンタテインメント エルエルシー | Dynamic music generation in gaming |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040111271A1 (en) * | 2001-12-10 | 2004-06-10 | Steve Tischer | Method and system for customizing voice translation of text to speech |
EP1577877A1 (en) * | 2002-10-24 | 2005-09-21 | National Institute of Advanced Industrial Science and Technology | Musical composition reproduction method and device, and method for detecting a representative motif section in musical composition data |
-
2012
- 2012-12-20 TW TW101148696A patent/TWI482149B/en not_active IP Right Cessation
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040111271A1 (en) * | 2001-12-10 | 2004-06-10 | Steve Tischer | Method and system for customizing voice translation of text to speech |
EP1577877A1 (en) * | 2002-10-24 | 2005-09-21 | National Institute of Advanced Industrial Science and Technology | Musical composition reproduction method and device, and method for detecting a representative motif section in musical composition data |
Also Published As
Publication number | Publication date |
---|---|
TW201426731A (en) | 2014-07-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Schuller | Speech emotion recognition: Two decades in a nutshell, benchmarks, and ongoing trends | |
Li et al. | Mec 2017: Multimodal emotion recognition challenge | |
Cunningham et al. | Supervised machine learning for audio emotion recognition: Enhancing film sound design using audio features, regression models and artificial neural networks | |
TWI486904B (en) | Method for rhythm visualization, system, and computer-readable memory | |
Yazdani et al. | Multimedia content analysis for emotional characterization of music video clips | |
CN107895577A (en) | Initiated using the task of long-tail voice command | |
Kabani et al. | Emotion based music player | |
CN113010138B (en) | Article voice playing method, device and equipment and computer readable storage medium | |
JP6535497B2 (en) | Music recommendation system, program and music recommendation method | |
Ottl et al. | Group-level speech emotion recognition utilising deep spectrum features | |
Drossos et al. | Affective acoustic ecology: Towards emotionally enhanced sound events | |
Wu et al. | Reducing model complexity for DNN based large-scale audio classification | |
Lee et al. | System for matching paintings with music based on emotions | |
Lemaitre et al. | Feelings elicited by auditory feedback from a computationally augmented artifact: The flops | |
Wang et al. | A review of sentiment semantic analysis technology and progress | |
WO2017143951A1 (en) | Expression feedback method and smart robot | |
Khurana et al. | Tri-integrated convolutional neural network for audio image classification using Mel-frequency spectrograms | |
Hou et al. | AI-based soundscape analysis: Jointly identifying sound sources and predicting annoyance | |
Haider et al. | An active data representation of videos for automatic scoring of oral presentation delivery skills and feedback generation | |
Schuller et al. | Three recent trends in paralinguistics on the way to omniscient machine intelligence | |
Wang et al. | Video affective content analysis by exploring domain knowledge | |
TWI486949B (en) | Music emotion classification method | |
TWI482149B (en) | The Method of Emotional Classification of Game Music | |
Cunha Sergio et al. | Emotional Video to Audio Transformation Using Deep Recurrent Neural Networks and a Neuro‐Fuzzy System | |
Huang et al. | Research on music emotion intelligent recognition and classification algorithm in music performance system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MM4A | Annulment or lapse of patent due to non-payment of fees |