WO2019227630A1 - 歌曲推荐方法和装置 - Google Patents

歌曲推荐方法和装置 Download PDF

Info

Publication number
WO2019227630A1
WO2019227630A1 PCT/CN2018/096331 CN2018096331W WO2019227630A1 WO 2019227630 A1 WO2019227630 A1 WO 2019227630A1 CN 2018096331 W CN2018096331 W CN 2018096331W WO 2019227630 A1 WO2019227630 A1 WO 2019227630A1
Authority
WO
WIPO (PCT)
Prior art keywords
song
type
target
emotion
current
Prior art date
Application number
PCT/CN2018/096331
Other languages
English (en)
French (fr)
Inventor
王杰
顾海倩
王姿雯
庄伯金
肖京
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2019227630A1 publication Critical patent/WO2019227630A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data

Definitions

  • the present application relates to the field of music recommendation systems, and more particularly, to a song recommendation method and device in the field of music recommendation systems.
  • the music recommendation system can recommend music to users based on their historical preferences when they listen to music.
  • Existing music recommendation systems usually implement strategies such as collaborative filtering based on user historical listening records to recommend music to users.
  • the existing music recommendation system simply refers to the user's historical listening record when making music recommendations, and the factors considered are relatively single; and for new users who do not have a historical listening record, they cannot provide effective music recommendations. Therefore, the user experience is poor.
  • This application provides a song recommendation method and device, which can flexibly recommend songs to users, thereby improving user experience.
  • this application provides a song recommendation method, which includes the following:
  • the song recommendation method in the process of performing song recommendation for the user, it is possible to flexibly recommend songs to the user in combination with the current mood of the user, thereby improving the user experience.
  • the present application further provides a song recommendation device, and the device specifically includes:
  • An obtaining unit configured to obtain at least one of text information, audio information, and image information of a user
  • a determining unit configured to determine a current emotion of the user according to the at least one piece of information and a mood analysis model obtained by the obtaining unit, and the mood analysis model is used to represent the at least one piece of information and the current mood Mapping relationship; determining as a target song type recommended by the user according to the current mood;
  • a recommendation unit is configured to recommend the at least one song to the user according to the target song type determined by the determining unit, and a song type of the at least one song belongs to the target song type.
  • the present application further provides a computer device, including a memory, a processor, a communication interface, and a computer program stored on the memory and executable on the processor, wherein the memory, the The processor and the communication interface communicate with each other through an internal connection path.
  • a computer device including a memory, a processor, a communication interface, and a computer program stored on the memory and executable on the processor, wherein the memory, the The processor and the communication interface communicate with each other through an internal connection path.
  • the present application also provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the steps of the foregoing method when executed by a processor.
  • FIG. 1 is a schematic flowchart of a song recommendation method according to an embodiment of the present application
  • FIG. 2 is a schematic flowchart of another song recommendation method according to an embodiment of the present application.
  • FIG. 3 is a schematic block diagram of a song recommendation device according to an embodiment of the present application.
  • FIG. 4 is a schematic block diagram of another song recommendation device according to an embodiment of the present application.
  • the existing music recommendation system recommends music to users through a collaborative filtering method based on the user's historical listening records.
  • the existing music recommendation system since the existing music recommendation system only refers to the historical listening record in the process of recommending music to the user, the angle of consideration is relatively single, and if the user is a new user, there is no relevant listening record, so it cannot be used for the user Recommended music. Therefore, the existing music recommendation system has poor flexibility in recommending music, and thus the user experience is poor.
  • This application provides a song recommendation method.
  • a song recommendation device obtains at least one of text information, audio information, and image information of a user; and determines a user's current emotion according to the at least one type of information and an emotion analysis model.
  • the sentiment analysis model is used to represent the mapping relationship between the at least one piece of information and the current sentiment; determine the target song type recommended by the user according to the current sentiment; and recommend the at least one song for the user according to the target song type.
  • the song type of at least one song belongs to the above target song type.
  • the song recommendation method provided in the embodiment of the present application can flexibly recommend songs to users, thereby improving user experience.
  • FIG. 1 shows a schematic flowchart of a song recommendation method 100 according to an embodiment of the present application.
  • S110 Acquire at least one piece of information of text information, audio information, and image information of the user.
  • S120 Determine a current emotion of the user according to the at least one information and emotion analysis model, where the emotion analysis model is used to represent a mapping relationship between the at least one information and the current emotion.
  • S130 Determine, according to the current mood, a target song type recommended by the user.
  • S140 Recommend the at least one song to the user according to the target song type, and the song type of the at least one song belongs to the target song type.
  • the method 100 may be performed by a song recommendation device.
  • the song recommendation device may be a device with a computing function.
  • the song recommendation device may be independent of a computer device, or may be integrated in a computer device and function as a function module in the computer device, which is not limited in the embodiment of the present application.
  • the text information of the user includes information used to represent the status vocabulary in the text data, where the status vocabulary can be divided into multiple types, including but not limited to: personal, money, health, and the like.
  • the status vocabulary information may include the frequency of occurrences of the status vocabulary.
  • the personal category information may include the frequency of the first person, the frequency of the second person, and the like.
  • the status vocabulary may be obtained from experience or may be obtained from machine learning, which is not limited in the embodiment of the present application.
  • the user's text data may include, for example, the user's Weibo text data, QQ text data, WeChat text data, short message text data, and other social-related text data.
  • the song recommendation device may obtain text data of the user, and determine the text information according to the text data.
  • the song recommendation device may obtain the text data in multiple ways, which is not limited in the embodiment of the present application.
  • the song recommendation device may use a web crawler technology to capture text data of a user in the Internet, or the song recommendation device may adopt a method of big data analysis, for example, spark technology, distributed computing ( (Hadoop) technology to obtain text data on a network data platform (cloud storage platform).
  • spark technology distributed computing ( (Hadoop) technology to obtain text data on a network data platform (cloud storage platform).
  • the song recommendation device collects chat records of a user on a QQ with a friend in a certain period of time.
  • the song recommendation device may obtain text data imported by a user.
  • the song recommendation device obtains a chat history with a friend on WeChat imported by the user.
  • the song recommendation device may obtain voice data of a user, and convert the voice data into text data.
  • the song recommendation device may obtain a call recording of a user, and perform text conversion on the call recording to obtain text data corresponding to the call recording.
  • the user's audio information includes information used to represent sound characteristics in the voice data.
  • the sound characteristics can be divided into multiple types, including but not limited to audio, rhythm, decibel, and frequency spectrum. Category, tone category, etc.
  • the song recommendation device may obtain voice data of the user, and determine the audio information according to the voice data.
  • the song recommendation device may obtain the voice data in multiple ways, which is not limited in the embodiment of the present application.
  • the song recommendation device may collect user's voice data through an audio collector.
  • the song recommendation device collects a user's call recording through an audio collector.
  • the song recommendation device collects a passage of a user expressing a mood or emotion through an audio collector.
  • the song recommendation device may obtain voice data imported by a user.
  • the song recommendation device acquires a segment of voice imported by a user.
  • the user's image information includes information used to represent facial features in the image data, where the facial features can be classified into multiple types, including but not limited to: eyes, noses, and mouths , Eyebrows and so on.
  • the eye-type information may include the position of the eye, the angle at which the corner of the eye is raised, and the like.
  • the song recommendation device may obtain image data of a user, and determine the image information according to the image data.
  • the song recommendation device may obtain the image data in multiple ways, which is not limited in the embodiment of the present application.
  • the song recommendation device may obtain the image data through a video collector.
  • the song recommendation device may collect a selfie image of a user through a video collector.
  • the song recommendation device may obtain image data imported by a user.
  • the song recommendation device may acquire pictures imported by a user that can reflect a mood or mood.
  • the song recommendation device may obtain video data of a user, and obtain image data according to the video data.
  • the song recommendation device collects a small video taken by a user through a video collector, and intercepts pictures in the small video.
  • the image information of the user may further include information for physical characteristics in the image data, or other information capable of reflecting the current mood of the user, which is not limited in the embodiment of the present application.
  • the song recommendation device obtains the text information, audio information, or image information.
  • the song recommendation device may also obtain the text information, audio information, or image information in other manners, such as ,
  • the staff directly inputs the user's text information, audio information, or image information into the song recommendation device, and the embodiment of the present application should not be limited to this.
  • the sentiment analysis model in S120 may include a first sentiment analysis model and a second sentiment analysis model.
  • the first sentiment analysis model is configured to indicate that each of the at least one type of information corresponds to each type of information. Mapping relationship between current emotions, and the second mood analysis model is used to represent the mapping relationship between the current emotions corresponding to each type of information and the current emotions of the user.
  • the current emotion corresponding to each type of information is determined according to each type of information in the at least one type of information and the first emotion analysis model; and the current emotion and second state information corresponding to each type of information are analyzed according to A model that determines the current mood of the user.
  • the current emotions in S120 may be divided into multiple types of emotions, and the multiple types of emotions include, but are not limited to, joy, sadness, pain, excitement, and tension.
  • each type of emotion may be divided into multiple intensity levels, and the current emotion may further include an intensity level of each type of emotion.
  • the current emotion may include a sadness level of 0.7, a happy level of 0, a pain level of 0.2, and a tense level of 0.1.
  • the sentiment analysis model may be pre-configured or established by the song recommendation device itself before the song recommendation device determines the current mood of the user based on the at least one information and sentiment analysis model. Examples do not limit this.
  • the song recommendation device may determine the target song type recommended by the user according to the current mood in multiple ways, which is not limited in this embodiment of the present application.
  • the song recommendation device may determine the target song type according to the current mood and the song recommendation model, and the song recommendation model is used to represent the current mood and the target song type. Mapping relationship.
  • the song recommendation device may determine the target song type according to the current emotion and an intensity level of the current emotion.
  • a song type that matches the current emotion is determined as the target song type, so that the user can achieve emotion and mood through the song Resonance.
  • the current emotion is a negative emotion
  • the intensity level of the negative emotion is greater than or equal to a preset second intensity level
  • a song type that matches an emotion opposite to the current emotion is determined as the target Song type, wherein the second intensity level is greater than the first intensity level, so that the user can achieve intervention and adjustment of the user's emotions through the song.
  • the song recommendation model may recommend the at least one song belonging to the target song type in the song library to the user according to the target song type and the song library, wherein the The song library includes multiple songs, the multiple songs belong to multiple song types, and the multiple song types include the target song type.
  • the song recommendation device may obtain the song library in advance.
  • the song library may be pre-configured or established by the song recommendation device, which is not limited in the embodiment of the present application.
  • the song recommendation device may obtain audio data of a first song in the song library; perform Fourier transform on the audio data to obtain the first song A Mel spectrum map; determining a song type of the first song according to the Mel spectrum map and a first song classification model, the first song classification model used to represent the Mel spectrum map and the song Mapping relationship between types; saving the first song and the song type of the first song in the song library.
  • the song recommendation device may obtain a lyrics text of a second song in the song library; and determine the first song according to the lyrics text and a second song classification model.
  • the second song classification model is used to represent the mapping relationship between the lyrics text and the song type; saving the second song and the song type of the second song to the song library in.
  • FIG. 2 is a schematic flowchart of another song recommendation method 200 according to an embodiment of the present application.
  • the method 200 may be performed by a song recommendation device, for example.
  • S210 Acquire at least one type of text information, audio information, and image information of the user.
  • S220 Determine the current emotion of the user according to the at least one information and emotion analysis model, where the emotion analysis model is used to represent a mapping relationship between the at least one information and the current emotion.
  • S240 Recommend the at least one song belonging to the target song type in the song library to the user according to the target song type and the song library, wherein the song library includes multiple songs, the Multiple songs belong to multiple song types, and the multiple song types include the target song type.
  • the song recommendation method provided by the embodiment of the present application is described above with reference to FIG. 1 and FIG. 2, and the song recommendation device provided by the embodiment of the present application is described below with reference to FIGS. 3 and 4.
  • FIG. 3 shows a schematic block diagram of a song recommendation device 300 according to an embodiment of the present application.
  • the device 300 includes:
  • An obtaining unit 310 configured to obtain at least one of text information, audio information, and image information of a user
  • a determining unit 320 is configured to determine a current emotion of the user according to the at least one type of information and the mood analysis model obtained by the obtaining unit, and the mood analysis model is used to represent the at least one kind of information and the current Mapping relationship of emotions; determining as a target song type recommended by the user according to the current emotions;
  • the recommending unit 330 is configured to recommend the at least one song to the user according to the target song type determined by the determining unit, and a song type of the at least one song belongs to the target song type.
  • determining the target song type recommended by the user according to the current mood includes: determining the target song type based on the current mood and a song recommendation model, where the song recommendation model is used to represent A mapping relationship between the current mood and the target song type.
  • the current emotion is divided into different intensity levels, and determining the target song type recommended by the user based on the current emotion includes: according to the current emotion and the intensity level of the current emotion, Determining the target song type.
  • determining the type of the target song recommended by the user according to the intensity level of the current emotion includes: when the intensity level of the current emotion is less than or equal to a preset first intensity level, The song type of the current emotion is determined as the target song type; or when the current emotion is a negative emotion and the intensity level of the negative emotion is greater than or equal to a preset second intensity level, it will meet the negative emotion.
  • the song type with the opposite mood is determined as the target song type, wherein the second intensity level is greater than the first intensity level.
  • recommending the at least one song for the user according to the target song type includes: recommending to the user that the song library belongs to the song according to the target song type and a song library.
  • the method before the recommending to the user at least one song belonging to the target song type in the song library according to the target song type and song library, includes: obtaining the Audio data of the first song in the song library; performing Fourier transform on the audio data to obtain a Mel spectrum map of the first song; determining the Mel spectrum map and the first song classification model according to the Mel spectrum map The song type of the first song, and the first song classification model is used to represent a mapping relationship between the Mel spectrum map and the song type.
  • the method before the recommending to the user at least one song belonging to the target song type in the song library according to the target song type and song library, includes: obtaining the The lyrics text of the second song in the song library; determining the song type of the second song according to the lyrics text and the second song classification model, the second song classification model is used to represent the lyrics text and the song Mapping relationship between types.
  • FIG. 4 shows a schematic block diagram of a song recommendation device 400 according to an embodiment of the present application.
  • the song recommendation device 400 may be the song recommendation device described in FIG. 4, and the song recommendation device may employ a hardware architecture as shown in FIG. 4.
  • the song recommendation device may include a processor 410, a communication interface 420, and a memory 430.
  • the processor 410, the communication interface 420, and the memory 430 communicate with each other through an internal connection path.
  • the related functions implemented by the determination unit 320 and the recommendation unit 330 in FIG. 3 may be implemented by the processor 410, and the related functions implemented by the acquisition unit 310 may be implemented by the processor 410 controlling the communication interface 420.
  • the processor 410 may include one or more processors, for example, one or more central processing units (CPUs).
  • CPUs central processing units
  • the processor may be a single-core CPU, or Can be a multi-core CPU.
  • the communication interface 420 is used for inputting and / or outputting data.
  • the communication interface may include a sending interface and a receiving interface.
  • the sending interface is used for outputting data and the receiving interface is used for inputting data.
  • the memory 430 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable memory (EPROM), read-only memory A compact disc (compact disc, read-only memory, CD-ROM).
  • RAM random access memory
  • ROM read-only memory
  • EPROM erasable programmable memory
  • read-only memory A compact disc (compact disc, read-only memory, CD-ROM).
  • the memory 430 is used to store related instructions and data.
  • the memory 430 is configured to store program codes and data of the song recommendation device, and may be a separate device or integrated in the processor 410.
  • the processor 410 is configured to control the communication interface to perform data transmission with other devices, such as a device or a song library that establishes a song library.
  • other devices such as a device or a song library that establishes a song library.
  • FIG. 4 only shows a simplified design of the song recommendation device.
  • the image retrieval device may also include other necessary components, including but not limited to any number of communication interfaces, processors, controllers, memories, etc., and all the song recommendation devices that can implement this application are included in this application. Within the scope of protection.
  • the song recommendation device 400 may be replaced with a chip device, for example, it may be a chip that can be used in the song recommendation device to implement the related functions of the processor 410 in the song recommendation device.
  • the chip device can be a field programmable gate array, a dedicated integrated chip, a system chip, a central processing unit, a network processor, a digital signal processing circuit, a microcontroller, and a programmable controller or other integrated chip to realize related functions.
  • the chip may optionally include one or more memories for storing program code, and when the code is executed, the processor implements a corresponding function.
  • the disclosed systems, devices, and methods may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each of the units may exist separately physically, or two or more units may be integrated into one unit.
  • the functions When the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially a part that contributes to the existing technology or a part of the technical solution can be embodied in the form of a software product.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
  • the foregoing storage medium includes various media that can store program codes, such as a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种歌曲推荐方法和装置,该方法包括:获取用户的文本信息、音频信息和图像信息中的至少一种信息(S110);根据该至少一种信息和情绪分析模型,确定该用户的当前情绪,该情绪分析模型用于表示该至少一种信息与该当前情绪的映射关系(S120);根据该当前情绪,确定为该用户推荐的目标歌曲类型(S130);根据该目标歌曲类型,为该用户推荐该至少一首歌曲,该至少一首歌曲的歌曲类型属于该目标歌曲类型(S140)。上述歌曲推荐方法和装置能够灵活为用户推荐歌曲,从而提高用户体验。

Description

歌曲推荐方法和装置
本申请申明享有2018年5月30日递交的申请号为CN 2018105379686、名称为“歌曲推荐方法和装置”的中国专利申请的优先权,该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本申请涉及音乐推荐系统领域,并且更具体地,涉及音乐推荐系统领域中的歌曲推荐方法和装置。
背景技术
音乐推荐系统能够在用户收听音乐的时候,根据用户的历史偏来为用户进行音乐推荐。现有的音乐推荐系统通常根据用户历史收听记录,采用协同过滤等策略实现对用户推荐音乐。
然而,现有的音乐推荐系统系统在进行音乐推荐的时候只是单纯的参考用户的历史收听记录,考虑的因素比较单一;并且对于不存在历史收听记录的新用户,不能提供有效的音乐推荐。因此,用户体验较差。
发明内容
本申请提供一种歌曲推荐方法和装置,能够灵活为用户推荐歌曲,从而提高用户体验。
为实现上述目的,本申请提供一种歌曲推荐方法,包括以下内容:
获取用户的文本信息、音频信息和图像信息中的至少一种信息;
根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;
根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
本申请实施例提供的歌曲推荐方法,在为用户进行歌曲推荐的过程中,能够结合用户的当前情绪,灵活为用户推荐歌曲,从而提高用户体验。
为实现上述目的,本申请还提供一种歌曲推荐装置,该装置具体包括:
获取单元,用于获取用户的文本信息、音频信息和图像信息中的至少一种信息;
确定单元,用于根据所述获取单元获取的所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
推荐单元,用于根据所述确定单元确定的所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
为实现上述目的,本申请还提供一种计算机设备,包括存储器、处理器、通信接口以及存储在所述存储器上并可在所述处理器上运行的计算机程序,其中,所述存储器、所述处理器以及所述通信接口之间通过内部连接通路互相通信,所述处理器执行所述计算机程序时实现上述方法的步骤。
为实现上述目的,本申请还提供计算机可读存储介质,其上存储有计算机程序,所述计算机程序被处理器执行时实现上述方法的步骤。
附图说明
图1是本申请实施例提供的歌曲推荐方法的示意性流程图;
图2是本申请实施例提供的另一歌曲推荐方法的示意性流程图;
图3是本申请实施例提供的歌曲推荐装置的示意性框图;
图4是本申请实施例提供的另一歌曲推荐装置的示意性框图。
具体实施方式
下面将结合附图,对本申请中的技术方案进行描述。
现有的音乐推荐系统根据用户的历史收听记录,通过协同过滤的方法为用户推荐音乐。
然而,由于现有的音乐推荐系统在对用户进行音乐推荐的过程中只参考历史收听记录,考虑的角度比较单一,并且如果该用户是新用户,没有相关的收听记录,这样就无法为该用户推荐音乐。因此,现有的音乐推荐系统推荐音乐的灵活性较差,从而用户体验较差。
本申请提供了一种歌曲推荐方法,歌曲推荐装置通过获取用户的文本信息、音频信息和图像信息中的至少一种信息;根据上述至少一种信息和情绪分析模型,确定用户的当前情绪,所述情绪分析模型用于表示上述至少一种信息与上述当前情绪的映射关系;根据当前情绪,确定为用户推荐的目标歌曲类型;根据上述目标歌曲类型,为用户推荐所述至少一首歌曲,上述至少一首歌曲的歌曲类型属于上述目标歌曲类型。采用本申请实施例提供的歌曲推荐方法,能够灵活为用户推荐歌曲,从而提高用户体验。
图1示出了本申请实施例提供的歌曲推荐方法100的示意性流程图。
S110,获取用户的文本信息、音频信息和图像信息中的至少一种信息。
S120,根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系。
S130,根据所述当前情绪,确定为所述用户推荐的目标歌曲类型。
S140,根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所 述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
可选地,该方法100可以由歌曲推荐装置执行,应理解,该歌曲推荐装置可以为具有计算功能的装置。
需要说明的是,该歌曲推荐装置可以独立于计算机设备,或者可以为集成于计算机设备中,作并作为该计算机设备中的功能模块,本申请实施例对此不作限定。
应理解,用户的文本信息包括用于表示文本数据中状态词汇的信息,其中,状态词汇可以分为多种类型,该多种类型包括但不限于:人称类、金钱类、健康类等。
例如,状态词汇的信息可以包括状态词汇出现的词频。
例如,人称类的信息可以包括第一人称的词频、第二人称的词频等。
可选地,该状态词汇可以为根据经验得到的或者可以为根据机器学习得到的,本申请实施例对此不作限定。
可选地,用户的文本数据例如可以包括该用户的微博文本数据、QQ文本数据、微信文本数据、短消息文本数据等与用户的社交相关的文本数据。
具体地,该歌曲推荐装置可以获取用户的文本数据,并根据该文本数据确定该文本信息。
可选地,该歌曲推荐装置可以通过多种方式获取该文本数据,本申请实施例对此不作限定。
在一种可能的实现方式中,该歌曲推荐装置可以采用网络爬虫技术抓取用户在互联网中的文本数据,或该歌曲推荐装置可以采用大数据分析的方法,例如,spark技术、分布式计算(Hadoop)技术等获取网络数据平台(云存储平台)上的文本数据。
例如,该歌曲推荐装置收集用户在某个时间段内与朋友在QQ上的聊天记录。
在另一种可能的实现方式中,该歌曲推荐装置可以获取用户导入的文本 数据。
例如,该歌曲推荐装置获取用户导入的与朋友在微信上的聊天记录。
在又一种可能的实现方式中,该歌曲推荐装置可以获取用户的语音数据,并将该语音转换为文本数据。
例如,该歌曲推荐装置可以获取用户的通话录音,通过对该通话录音进行文字转换,得到与该通话录音对应的文字数据。
应理解,用户的音频信息包括用于表示语音数据中声音特征的信息,其中,该声音特征可以分为多种类型,该多种类型包括但不限于音频类、节奏类、分贝类、频谱类、音色类等。
具体地,该歌曲推荐装置可以获取用户的语音数据,并根据该语音数据确定该音频信息。
可选地,该歌曲推荐装置可以通过多种方式获取该语音数据,本申请实施例对此不作限定。
在一种可能的实现方式中,该歌曲推荐装置可以通过音频采集器采集用户的语音数据。
例如,该歌曲推荐装置通过音频采集器采集用户的通话录音。
又例如,该歌曲推荐装置通过音频采集器采集用户表达心情或情绪的一段话。
在另一种可能的实现方式中,该歌曲推荐装置可以获取用户导入的语音数据。
例如,该歌曲推荐装置获取用户导入的一段语音。
应理解,用户的图像信息包括用于表示图像数据中人脸特征的信息,其中,该人脸特征可以分为多种类型,该多种类型包括但不限于:眼睛类、鼻子类、嘴巴类、眉毛类等。
例如,眼睛类的信息可以包括眼睛的位置,眼角上扬的角度等。
具体地,该歌曲推荐装置可以获取用户的图像数据,并根据该图像数据 确定该图像信息。
可选地,该歌曲推荐装置可以通过多种方式获取该图像数据,本申请实施例对此不作限定。
在一种可能的实现方式中,该歌曲推荐装置可以通过视频采集器获取该图像数据。
例如,该歌曲推荐装置可以通过视频采集器采集用户的自拍图像。
在另一种可能的实现方式中,该歌曲推荐装置可以获取用户导入的图像数据。
例如,该歌曲推荐装置可以获取用户导入的能够反映心情或情绪的图片。
在又一种可能的实现方式中,该歌曲推荐装置可以获取用户的视频数据,根据该视频数据,获取图像数据。
例如,该歌曲推荐装置通过视频采集器采集用户拍摄自己的小视频,并截取该小视频中的图片。
可选地,该用户的图像信息还可以包括用于图像数据中肢体特征的信息,或其他能够反映用户的当前情绪的信息,本申请实施例对此不作限定。
应理解,上面示例性地列举出该歌曲推荐装置获取该文本信息、音频信息或图像信息的可能的实现方式,该歌曲推荐装置还可以通过其他方式获取该文本信息、音频信息或图像信息,例如,工作人员直接将用户的文本信息、音频信息或图像信息输入该歌曲推荐装置等,本申请实施例不应受限于此。
可选地,S120中的情绪分析模型可以包括第一情绪分析模型和第二情绪分析模型,该第一情绪分析模型用于表示该至少一种信息中的每种信息与该每种信息对应的当前情绪的映射关系,该第二情绪分析模型用于表示该每种信息对应的当前情绪和该用户的当前情绪的映射关系。
相应地,S120中,根据该至少一种信息中的每种信息和第一情绪分析模型,确定该每种信息对应的当前情绪;并根据该每种信息对应的当前情绪和第二状态信息分析模型,确定该用户的当前情绪。
可选地,S120中的当前情绪可以分为多种类型的情绪,该多种类型的情绪包括但不限于:高兴、悲伤、痛苦、兴奋、紧张等。
可选地,该每种类型的情绪可以分为多个强度等级,该当前情绪还可以包括每种类型的情绪的强度等级。
例如,该当前情绪可以包括悲伤的强度等级为0.7,高兴的强度等级为0,痛苦的强度等级为0.2,紧张的强度等级为0.1。
可选地,该情绪分析模型可以为预配置的,或者在该歌曲推荐装置根据该至少一种信息和情绪分析模型,确定该用户的当前情绪之前,该歌曲推荐装置自己建立的,本申请实施例对此不作限定。
可选地,S130中,该歌曲推荐装置可以通过多种方式根据该当前情绪,确定为该用户推荐的目标歌曲类型,本申请实施例对此不作限定。
在一种可能的实现方式中,该歌曲推荐装置可以根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
在另一种可能的实现方式中,该歌曲推荐装置可以根据所述当前情绪和所述当前情绪的强度等级,确定所述目标歌曲类型。
例如,当所述当前情绪的强度等级小于或等于预设的第一强度等级时,将符合所述当前情绪的歌曲类型确定为所述目标歌曲类型,以使得用户能够通过歌曲达到情绪和心情上的共鸣。
又例如,当所述当前情绪为消极情绪,且所述消极情绪的强度等级大于或等于预设的第二强度等级时,将符合与所述当前情绪相反的情绪的歌曲类型确定为所述目标歌曲类型,其中,所述第二强度等级大于所述第一强度等级,以使得用户能够通过歌曲达到对用户情绪的干预和调节。
可选地,S140中,该歌曲推荐模型可以根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述 多种歌曲类型包括所述目标歌曲类型。
可选地,在该歌曲推荐装置根据所述目标歌曲类型和歌曲库,为所述用户推荐所述至少一首歌曲之前,该歌曲推荐装置可以预先获取该歌曲库。
可选地,该歌曲库可以为预配置的或该歌曲推荐装置建立的,本申请实施例对此不作限定。
在一种可能的实现方式中,对于没有歌词的歌曲,该歌曲推荐装置可以获取所述歌曲库中第一歌曲的音频数据;对所述音频数据进行傅里叶变换,得到所述第一歌曲的梅尔频谱图;根据所述梅尔频谱图和第一歌曲分类模型,确定所述第一歌曲的歌曲类型,所述第一歌曲分类模型用于表示所述梅尔频谱图与所述歌曲类型之间的映射关系;将该第一歌曲和该第一歌曲的歌曲类型保存在所述歌曲库中。
在另一种可能的实现方式中,对于有歌词的歌曲,该歌曲推荐装置可以获取所述歌曲库中第二歌曲的歌词文本;根据所述歌词文本和第二歌曲分类模型,确定所述第二歌曲的歌曲类型,所述第二歌曲分类模型用于表示所述歌词文本与所述歌曲类型之间的映射关系;将该第二歌曲和该第二歌曲的歌曲类型保存至所述歌曲库中。
图2示出了本申请实施例提供的另一歌曲推荐方法200的示意性流程图,该方法200例如可以由歌曲推荐装置执行。
S210,获取用户的文本信息、音频信息和图像信息中的至少一种信息。
S220,根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系。
S230,根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
S240,根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中 属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述多种歌曲类型包括所述目标歌曲类型。
上面结合图1和图2介绍了本申请实施例提供的歌曲推荐方法,下面将结合图3和图4介绍本申请实施例提供的歌曲推荐装置。
图3示出了本申请实施例提供的歌曲推荐装置300的示意性框图。该装置300包括:
获取单元310,用于获取用户的文本信息、音频信息和图像信息中的至少一种信息;
确定单元320,用于根据所述获取单元获取的所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
推荐单元330,用于根据所述确定单元确定的所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
可选地,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
可选地,所述当前情绪分为不同的强度等级,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:根据所述当前情绪和所述当前情绪的强度等级,确定所述目标歌曲类型。
可选地,所述根据所述当前情绪的强度等级,确定为所述用户推荐的目标歌曲类型,包括:当所述当前情绪的强度等级小于或等于预设的第一强度等级时,将符合所述当前情绪的歌曲类型确定为所述目标歌曲类型;或当所述当前情绪为消极情绪,且该消极情绪的强度等级大于或等于预设的第二强 度等级时,将符合与所述消极情绪相反的情绪的歌曲类型确定为所述目标歌曲类型,其中,所述第二强度等级大于所述第一强度等级。
可选地,所述根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,包括:根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述多种歌曲类型包括所述目标歌曲类型。
可选地,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:获取所述歌曲库中第一歌曲的音频数据;对所述音频数据进行傅里叶变换,得到所述第一歌曲的梅尔频谱图;根据所述梅尔频谱图和第一歌曲分类模型,确定所述第一歌曲的歌曲类型,所述第一歌曲分类模型用于表示所述梅尔频谱图与所述歌曲类型之间的映射关系。
可选地,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:获取所述歌曲库中第二歌曲的歌词文本;根据所述歌词文本和第二歌曲分类模型,确定所述第二歌曲的歌曲类型,所述第二歌曲分类模型用于表示所述歌词文本与所述歌曲类型之间的映射关系。
图4示出了本申请实施例提供的歌曲推荐装置400的示意性框图。该歌曲推荐装置400可以为图4中所述的歌曲推荐装置,该歌曲推荐装置可以采用如图4所示的硬件架构。该歌曲推荐装置可以包括处理器410、通信接口420和存储器430,该处理器410、通信接口420和存储器430通过内部连接通路互相通信。图3中的确定单元320和推荐单元330所实现的相关功能可以由处理器410来实现,获取单元310所实现的相关功能可以由处理器410控制通信接口420来实现。
该处理器410可以包括是一个或多个处理器,例如包括一个或多个中央 处理单元(central processing unit,CPU),在处理器是一个CPU的情况下,该CPU可以是单核CPU,也可以是多核CPU。
该通信接口420用于输入和/或输出数据。该通信接口可以包括发送接口和接收接口,发送接口用于输出数据,接收接口用于输入数据。
该存储器430包括但不限于是随机存取存储器(random access memory,RAM)、只读存储器(read-only memory,ROM)、可擦除可编程存储器(erasable programmable read only memory,EPROM)、只读光盘(compact disc read-only memory,CD-ROM),该存储器430用于存储相关指令及数据。
存储器430用于存储歌曲推荐装置的程序代码和数据,可以为单独的器件或集成在处理器410中。
具体地,所述处理器410用于控制通信接口与其它装置,例如与建立歌曲库的装置或歌曲库进行数据传输。具体可参见方法实施例中的描述,在此不再赘述。
可以理解的是,图4仅仅示出了歌曲推荐装置的简化设计。在实际应用中,图像检索装置还可以分别包含必要的其他元件,包含但不限于任意数量的通信接口、处理器、控制器、存储器等,而所有可以实现本申请的歌曲推荐装置都在本申请的保护范围之内。
在一种可能的设计中,歌曲推荐装置400可以被替换为芯片装置,例如可以为可用于歌曲推荐装置中的芯片,用于实现歌曲推荐装置中处理器410的相关功能。该芯片装置可以为实现相关功能的现场可编程门阵列,专用集成芯片,系统芯片,中央处理器,网络处理器,数字信号处理电路,微控制器,还可以采用可编程控制器或其他集成芯片。该芯片中,可选的可以包括一个或多个存储器,用于存储程序代码,当所述代码被执行时,使得处理器实现相应的功能。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结 合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的 存储介质包括:U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应以所述权利要求的保护范围为准。

Claims (20)

  1. 一种歌曲推荐方法,其特征在于,包括:
    获取用户的文本信息、音频信息和图像信息中的至少一种信息;
    根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;
    根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
    根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
  3. 根据权利要求1所述的方法,其特征在于,所述当前情绪分为不同的强度等级,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和所述当前情绪的强度等级,确定所述目标歌曲类型。
  4. 根据权利要求3所述的方法,其特征在于,所述根据所述当前情绪的强度等级,确定为所述用户推荐的目标歌曲类型,包括:
    当所述当前情绪的强度等级小于或等于预设的第一强度等级时,将符合所述当前情绪的歌曲类型确定为所述目标歌曲类型;或
    当所述当前情绪为消极情绪,且该消极情绪的强度等级大于或等于预设的第二强度等级时,将符合与所述消极情绪相反的情绪的歌曲类型确定为所述目标歌曲类型,其中,所述第二强度等级大于所述第一强度等级。
  5. 根据权利要求1至4中任一项所述的方法,其特征在于,所述根据所 述目标歌曲类型,为所述用户推荐所述至少一首歌曲,包括:
    根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述多种歌曲类型包括所述目标歌曲类型。
  6. 根据权利要求5所述的方法,其特征在于,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:
    获取所述歌曲库中第一歌曲的音频数据;
    对所述音频数据进行傅里叶变换,得到所述第一歌曲的梅尔频谱图;
    根据所述梅尔频谱图和第一歌曲分类模型,确定所述第一歌曲的歌曲类型,所述第一歌曲分类模型用于表示所述梅尔频谱图与所述歌曲类型之间的映射关系。
  7. 根据权利要求5所述的方法,其特征在于,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:
    获取所述歌曲库中第二歌曲的歌词文本;
    根据所述歌词文本和第二歌曲分类模型,确定所述第二歌曲的歌曲类型,所述第二歌曲分类模型用于表示所述歌词文本与所述歌曲类型之间的映射关系。
  8. 一种歌曲推荐装置,其特征在于,包括:
    获取单元,用于获取用户的文本信息、音频信息和图像信息中的至少一种信息;
    确定单元,用于根据所述获取单元获取的所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
    推荐单元,用于根据所述确定单元确定的所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
  9. 一种计算机设备,包括存储器、处理器、通信接口以及存储在所述存储器上并可在所述处理器上运行的计算机程序,其中,所述存储器、所述处理器以及所述通信接口之间通过内部连接通路互相通信,其特征在于,所述处理器执行所述计算机程序时实现歌曲推荐方法的以下步骤:
    获取用户的文本信息、音频信息和图像信息中的至少一种信息;
    根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;
    根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
    根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
  10. 根据权利要求9所述的计算机设备,其特征在于,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
  11. 根据权利要求9所述的计算机设备,其特征在于,所述当前情绪分为不同的强度等级,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和所述当前情绪的强度等级,确定所述目标歌曲类型。
  12. 根据权利要求11所述的计算机设备,其特征在于,所述根据所述当前情绪的强度等级,确定为所述用户推荐的目标歌曲类型,包括:
    当所述当前情绪的强度等级小于或等于预设的第一强度等级时,将符合所述当前情绪的歌曲类型确定为所述目标歌曲类型;或
    当所述当前情绪为消极情绪,且该消极情绪的强度等级大于或等于预设 的第二强度等级时,将符合与所述消极情绪相反的情绪的歌曲类型确定为所述目标歌曲类型,其中,所述第二强度等级大于所述第一强度等级。
  13. 根据权利要求9至12中任一项所述的计算机设备,其特征在于,所述根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,包括:
    根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述多种歌曲类型包括所述目标歌曲类型。
  14. 根据权利要求13所述的计算机设备,其特征在于,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:
    获取所述歌曲库中第一歌曲的音频数据;
    对所述音频数据进行傅里叶变换,得到所述第一歌曲的梅尔频谱图;
    根据所述梅尔频谱图和第一歌曲分类模型,确定所述第一歌曲的歌曲类型,所述第一歌曲分类模型用于表示所述梅尔频谱图与所述歌曲类型之间的映射关系。
  15. 根据权利要求13所述的计算机设备,其特征在于,在所述根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲之前,所述方法包括:
    获取所述歌曲库中第二歌曲的歌词文本;
    根据所述歌词文本和第二歌曲分类模型,确定所述第二歌曲的歌曲类型,所述第二歌曲分类模型用于表示所述歌词文本与所述歌曲类型之间的映射关系。
  16. 一种计算机可读存储介质,用于存储计算机程序,其特征在于,所述计算机程序被处理器执行时实现歌曲推荐方法的以下步骤:
    获取用户的文本信息、音频信息和图像信息中的至少一种信息;
    根据所述至少一种信息和情绪分析模型,确定所述用户的当前情绪,所 述情绪分析模型用于表示所述至少一种信息与所述当前情绪的映射关系;
    根据所述当前情绪,确定为所述用户推荐的目标歌曲类型;
    根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,所述至少一首歌曲的歌曲类型属于所述目标歌曲类型。
  17. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和歌曲推荐模型,确定所述目标歌曲类型,所述歌曲推荐模型用于表示所述当前情绪和所述目标歌曲类型之间的映射关系。
  18. 根据权利要求16所述的计算机可读存储介质,其特征在于,所述当前情绪分为不同的强度等级,所述根据所述当前情绪,确定为所述用户推荐的目标歌曲类型,包括:
    根据所述当前情绪和所述当前情绪的强度等级,确定所述目标歌曲类型。
  19. 根据权利要求18所述的计算机可读存储介质,其特征在于,所述根据所述当前情绪的强度等级,确定为所述用户推荐的目标歌曲类型,包括:
    当所述当前情绪的强度等级小于或等于预设的第一强度等级时,将符合所述当前情绪的歌曲类型确定为所述目标歌曲类型;或
    当所述当前情绪为消极情绪,且该消极情绪的强度等级大于或等于预设的第二强度等级时,将符合与所述消极情绪相反的情绪的歌曲类型确定为所述目标歌曲类型,其中,所述第二强度等级大于所述第一强度等级。
  20. 根据权利要求16至19中任一项所述的计算机可读存储介质,其特征在于,所述根据所述目标歌曲类型,为所述用户推荐所述至少一首歌曲,包括:
    根据所述目标歌曲类型和歌曲库,为所述用户推荐所述歌曲库中属于所述目标歌曲类型的所述至少一首歌曲,其中,所述歌曲库中包括多首歌曲,所述多首歌曲属于多种歌曲类型,所述多种歌曲类型包括所述目标歌曲类型。
PCT/CN2018/096331 2018-05-30 2018-07-19 歌曲推荐方法和装置 WO2019227630A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810537968.6A CN108804609A (zh) 2018-05-30 2018-05-30 歌曲推荐方法和装置
CN201810537968.6 2018-05-30

Publications (1)

Publication Number Publication Date
WO2019227630A1 true WO2019227630A1 (zh) 2019-12-05

Family

ID=64089368

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/096331 WO2019227630A1 (zh) 2018-05-30 2018-07-19 歌曲推荐方法和装置

Country Status (2)

Country Link
CN (1) CN108804609A (zh)
WO (1) WO2019227630A1 (zh)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109684501B (zh) * 2018-11-26 2023-08-22 平安科技(深圳)有限公司 歌词信息生成方法及其装置
CN111583890A (zh) * 2019-02-15 2020-08-25 阿里巴巴集团控股有限公司 音频分类方法和装置
CN110473546B (zh) * 2019-07-08 2022-05-31 华为技术有限公司 一种媒体文件推荐方法及装置
CN111128103A (zh) * 2019-12-19 2020-05-08 北京凯来科技有限公司 一种沉浸式ktv智能点唱系统
CN111737414B (zh) * 2020-06-04 2024-08-09 腾讯音乐娱乐科技(深圳)有限公司 一种歌曲推荐方法及装置、服务器、存储介质
CN111753049B (zh) * 2020-06-15 2024-04-16 广东美的厨房电器制造有限公司 菜谱推荐方法、装置、家电设备及存储介质
CN112905835B (zh) * 2021-02-26 2022-11-11 成都潜在人工智能科技有限公司 一种多模态乐曲标题生成方法、装置及存储介质
CN113010728A (zh) * 2021-04-06 2021-06-22 金宝贝网络科技(苏州)有限公司 一种歌曲推荐方法、系统、智能设备及存储介质
CN113572893B (zh) * 2021-07-13 2023-03-14 青岛海信移动通信技术股份有限公司 一种终端设备、情绪反馈方法及存储介质
CN113518263B (zh) * 2021-07-23 2023-02-10 南京炫佳网络科技有限公司 交互式网络电视的视频推荐方法、装置、电视及存储介质
CN113744445B (zh) * 2021-09-06 2024-01-05 北京雷石天地电子技术有限公司 一种比赛投票方法、装置、计算机设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103412646A (zh) * 2013-08-07 2013-11-27 南京师范大学 基于脑机交互的音乐情绪化推荐方法
CN105426381A (zh) * 2015-08-27 2016-03-23 浙江大学 一种基于微博情绪上下文的音乐推荐方法
CN106202103A (zh) * 2015-05-06 2016-12-07 阿里巴巴集团控股有限公司 音乐推荐方法和装置
CN107562850A (zh) * 2017-08-28 2018-01-09 百度在线网络技术(北京)有限公司 音乐推荐方法、装置、设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103412646A (zh) * 2013-08-07 2013-11-27 南京师范大学 基于脑机交互的音乐情绪化推荐方法
CN106202103A (zh) * 2015-05-06 2016-12-07 阿里巴巴集团控股有限公司 音乐推荐方法和装置
CN105426381A (zh) * 2015-08-27 2016-03-23 浙江大学 一种基于微博情绪上下文的音乐推荐方法
CN107562850A (zh) * 2017-08-28 2018-01-09 百度在线网络技术(北京)有限公司 音乐推荐方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN108804609A (zh) 2018-11-13

Similar Documents

Publication Publication Date Title
WO2019227630A1 (zh) 歌曲推荐方法和装置
US11430439B2 (en) System and method for providing assistance in a live conversation
US11763811B2 (en) Oral communication device and computing system for processing data and outputting user feedback, and related methods
KR102331049B1 (ko) 통신 개시를 위한 사용자 신호 레버리징
US10692606B2 (en) Stress level reduction using haptic feedback
US11228804B2 (en) Identification and instantiation of community driven content
US11792141B2 (en) Automated messaging reply-to
US11197074B2 (en) Synchronized video annotation and control system for sexual stimulation devices
US11928985B2 (en) Content pre-personalization using biometric data
CN110060656B (zh) 模型管理和语音合成方法、装置和系统及存储介质
EP3267333A1 (en) Local processing of biometric data for a content selection system
US10638174B2 (en) Synchronized video control system for sexual stimulation devices
US20200175056A1 (en) Digital content delivery based on predicted effect
WO2019227633A1 (zh) 建立用户画像和建立状态信息分析模型的方法和装置
WO2023045912A1 (en) Selective content transfer for streaming content
US20200112755A1 (en) Providing relevant and authentic channel content to users based on user persona and interest
WO2022141142A1 (zh) 一种确定目标音视频的方法及系统
CN114115533A (zh) 智能交互方法和装置
CN113573128A (zh) 一种音频处理方法、装置、终端以及存储介质
US10296723B2 (en) Managing companionship data
US11552966B2 (en) Generating and mutually maturing a knowledge corpus
US11630865B2 (en) User reaction based information options
CN117319340A (zh) 语音消息的播放方法、装置、终端及存储介质
CN111610851A (zh) 互动方法、装置以及用于实现该互动方法的用户终端
JP2021193469A (ja) 情報処理システム、情報処理方法、および記録媒体

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18920887

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18920887

Country of ref document: EP

Kind code of ref document: A1