WO2022002204A1 - 辅助单车训练的方法及装置、网络模型的训练方法及装置 - Google Patents

辅助单车训练的方法及装置、网络模型的训练方法及装置 Download PDF

Info

Publication number
WO2022002204A1
WO2022002204A1 PCT/CN2021/104029 CN2021104029W WO2022002204A1 WO 2022002204 A1 WO2022002204 A1 WO 2022002204A1 CN 2021104029 W CN2021104029 W CN 2021104029W WO 2022002204 A1 WO2022002204 A1 WO 2022002204A1
Authority
WO
WIPO (PCT)
Prior art keywords
audio
training
user
processed
information
Prior art date
Application number
PCT/CN2021/104029
Other languages
English (en)
French (fr)
Inventor
陈骋
Original Assignee
随机漫步(上海)体育科技有限公司
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 随机漫步(上海)体育科技有限公司 filed Critical 随机漫步(上海)体育科技有限公司
Publication of WO2022002204A1 publication Critical patent/WO2022002204A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B22/00—Exercising apparatus specially adapted for conditioning the cardio-vascular system, for training agility or co-ordination of movements
    • A63B22/06—Exercising apparatus specially adapted for conditioning the cardio-vascular system, for training agility or co-ordination of movements with support elements performing a rotating cycling movement, i.e. a closed path movement
    • A63B22/0605—Exercising apparatus specially adapted for conditioning the cardio-vascular system, for training agility or co-ordination of movements with support elements performing a rotating cycling movement, i.e. a closed path movement performing a circular movement, e.g. ergometers
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
    • G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use

Definitions

  • the present application relates to the technical field of signal processing, and in particular, to a method and device for assisting bicycle training, a method and device for training a network model, a computer-readable storage medium, and an electronic device.
  • training sessions in existing spinning bikes are pre-recorded by trainers. Specifically, any training session, whether the session mode or the session music, is pre-set by the coach. Therefore, the existing training courses of spinning bikes cannot meet the personalized training needs of users, and the user experience favorability is extremely poor.
  • Embodiments of the present application provide a method and apparatus for assisting bicycle training, a method and apparatus for training a network model, a computer-readable storage medium, and an electronic device.
  • an embodiment of the present application provides a method for assisting bicycle training, and the method for assisting bicycle training includes: determining the to-be-processed audio corresponding to the first user; inputting the to-be-processed audio into an audio splitting model to generate Processing audio element information corresponding to the audio; generating motion data corresponding to the audio to be processed based on the audio element information, where the motion data is motion data used to assist the first user in cycling training.
  • an embodiment of the present application provides a method for training a network model.
  • the method for training a network model includes: determining training audio and audio element information corresponding to the training audio; establishing an initial network model, and based on the training audio and audio
  • the element information trains an initial network model to generate an audio splitting model, wherein the audio splitting model is used to generate audio element information corresponding to the to-be-processed audio based on the to-be-processed audio.
  • an embodiment of the present application provides a device for assisting bicycle training, and the device for assisting bicycle training includes: a to-be-processed audio determination module for determining the to-be-processed audio corresponding to the first user; a first generation module, For inputting the audio to be processed into the audio splitting model, to generate audio element information corresponding to the audio to be processed; the second generation module is used to generate motion data corresponding to the audio to be processed based on the audio element information, wherein the motion data is used for Motion data for assisting the first user in cycling training.
  • an embodiment of the present application provides a training device for a network model.
  • the training device for a network model includes: a determination module for determining training audio and audio element information corresponding to the training audio; a training module for establishing The initial network model is trained based on the training audio and the audio element information to generate an audio splitting model, wherein the audio splitting model is used to generate audio element information corresponding to the to-be-processed audio based on the to-be-processed audio.
  • an embodiment of the present application provides a computer-readable storage medium, where the storage medium stores a computer program, and the computer program is used to execute the method for assisting bicycle training described in the foregoing embodiments, or to Execute the training method of the network model described in the above embodiment.
  • an embodiment of the present application provides an electronic device, the electronic device includes: a processor; a memory for storing instructions executable by the processor; the processor for executing the The method for assisting bicycle training described above, or the training method for implementing the network model described in the above embodiments.
  • an embodiment of the present application provides a bicycle, and the bicycle is loaded with the apparatus for assisting bicycle training and/or the training apparatus for a network model described in the above embodiments.
  • the embodiment of the present application does not need to generate a training course in advance, and defines the exercise data and the audio corresponding to the exercise data in the training course, which can assist the training.
  • the method for assisting bicycle training provided by the embodiments of the present application can generate motion data for assisting the first user in bicycle training based on the to-be-processed audio corresponding to the first user, thereby meeting the personalized training needs of the first user and improving user experience Favorability.
  • the motion data is determined based on the audio element information generated by the audio splitting model, the matching degree between the motion data and the audio to be processed is higher. Therefore, the embodiment of the present application can also further improve the training effect.
  • FIG. 1 is a schematic diagram of a scene to which the embodiment of the present application is applied.
  • FIG. 2 is a schematic flowchart of a method for assisting bicycle training provided by an exemplary embodiment of the present application.
  • FIG. 3 shows a schematic flowchart of generating motion data corresponding to audio to be processed based on audio element information according to an exemplary embodiment of the present application.
  • FIG. 4 shows a schematic diagram of an actual generation process of motion data provided by an exemplary embodiment of the present application.
  • FIG. 5 is a schematic flowchart of a method for assisting bicycle training provided by another exemplary embodiment of the present application.
  • Fig. 6 is a schematic flowchart of a method for assisting bicycle training provided by yet another exemplary embodiment of the present application.
  • FIG. 7 shows a schematic flowchart of determining the audio to be processed corresponding to the first user according to an exemplary embodiment of the present application.
  • FIG. 8 is a schematic flowchart of a training method for a network model provided by an exemplary embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of an apparatus for assisting bicycle training provided by an exemplary embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a second generation module provided by an exemplary embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of an apparatus for assisting bicycle training provided by another exemplary embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of an apparatus for assisting bicycle training provided by yet another exemplary embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of a to-be-processed audio determination module provided by an exemplary embodiment of the present application.
  • FIG. 14 is a schematic structural diagram of an apparatus for training a network model according to an exemplary embodiment of the present application.
  • FIG. 15 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present application.
  • the bicycles in the embodiments of the present application may be ordinary bicycles used for outdoor cycling, or may be fitness equipment used in indoor bicycle training courses, which are not limited in the embodiments of the present application.
  • the user terminal in this embodiment of the present application may be a user terminal disposed on a bicycle, or may be a mobile terminal, such as a mobile phone or a tablet computer.
  • FIG. 1 is a schematic diagram of a scene to which the embodiment of the present application is applied.
  • the applicable scene of the embodiment of the present application includes a bicycle 110 and a server 120 , wherein a user terminal 111 is loaded on the bicycle 110 , and a communication connection relationship exists between the server 120 and the user terminal 111 .
  • the user terminal 111 is configured to acquire relevant information of the first user, and implement information interaction with the server 120 based on the acquired relevant information.
  • the server 120 stores data such as an audio split model.
  • the server 120 determines the to-be-processed audio corresponding to the first user based on the user terminal 111 in the bicycle 110 , and then the server 120 inputs the acquired to-be-processed audio into the audio splitting model to generate a corresponding to-be-processed audio
  • the audio element information based on the audio element information, generates motion data corresponding to the audio to be processed, and transmits the motion data to the user terminal 111 to assist the first user in cycling training. That is, this scenario implements a method to assist bicycle training.
  • the bicycle 110 further includes a sensor that is communicatively connected to the user terminal 111 .
  • the sensor is used to acquire the sports performance data of the first user, so that the user terminal 111 or the server 120 can perform operations such as sports evaluation based on the acquired sports performance data.
  • the sensor is arranged in the pedal of the bicycle 110, and a lot of motion information such as pedaling strength, pedaling frequency, and pedaling time point of the first user is collected by means of the sensor provided in the pedal.
  • the senor and the user terminal 111 establish a communication connection relationship based on the Bluetooth technology.
  • FIG. 2 is a schematic flowchart of a method for assisting bicycle training provided by an exemplary embodiment of the present application. As shown in FIG. 2 , the method for assisting bicycle training provided by the embodiment of the present application includes the following steps.
  • Step S210 determining the to-be-processed audio corresponding to the first user.
  • the first user is a user who wants to perform cycling training by means of a bicycle.
  • the to-be-processed audio refers to the to-be-processed audio corresponding to the audio voiceprint information input by the first user.
  • Step S220 Input the audio to be processed into the audio splitting model to generate audio element information corresponding to the audio to be processed.
  • the audio element information includes at least one of rhythm information, tempo information, and energy information.
  • the audio splitting model is a deep learning-based neural network model, such as a convolutional neural network model including a convolutional layer and other structures.
  • Step S230 generating motion data corresponding to the audio to be processed based on the audio element information.
  • the motion data mentioned in step S230 is motion data used to assist the first user in cycling training.
  • the motion data includes at least one of cadence data, speed data, and cadence data.
  • the motion data further includes at least one of score data, difficulty rating data, highest score data, and segment score data.
  • the audio element information can better characterize the audio characteristics of the audio to be processed, motion data corresponding to the audio to be processed can be more accurately generated based on the audio element information, thereby matching more suitable motion data for the audio to be processed.
  • the audio characteristics include information such as audio style, audio type, and audio climax area.
  • the embodiment of the present application does not need to generate a training course in advance, and defines the exercise data and the audio corresponding to the exercise data in the training course, which can assist the training.
  • the method for assisting bicycle training provided by the embodiments of the present application can generate motion data for assisting the first user in bicycle training based on the to-be-processed audio corresponding to the first user, thereby meeting the personalized training needs of the first user and improving user experience Favorability.
  • the motion data is determined based on the audio element information generated by the audio splitting model, the matching degree between the motion data and the audio to be processed is higher. Therefore, the embodiment of the present application can also further improve the training effect.
  • FIG. 3 shows a schematic flowchart of generating motion data corresponding to audio to be processed based on audio element information according to an exemplary embodiment of the present application.
  • the embodiment shown in FIG. 3 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 3 and the embodiment shown in FIG. 2 , and the similarities will not be repeated. .
  • the step of generating motion data corresponding to the audio to be processed based on the audio element information includes the following steps.
  • Step S231 determining the historical training data corresponding to the first user.
  • the historical training data can represent information such as the first user's exercise ability and exercise preference. Then, using the historical training data as one of the reference parameters for generating the exercise data can further improve the first user's satisfaction with the generated exercise data.
  • the historical training data includes at least one of historical course score information, historical course matching curve information, historical course participation duration information, and historical training time information.
  • Step S232 using a preset data generation algorithm to generate motion data based on historical training data and audio element information.
  • the preset data generation algorithm mentioned in step S232 refers to an algorithm capable of generating motion data by integrating historical training data and audio element information.
  • the audio element information includes beat information of the audio to be processed
  • the motion data includes cadence data
  • the cadence data specifically includes a first-intensity cadence, a second-intensity cadence, and a third-intensity cadence. If it should be determined based on the beat information that the cadence data of the exercise data is the cadence of the second intensity, and it is found that the historical training courses of the first user are all the cadence of the third intensity according to the historical training data of the first user, then the preset data is generated. After processing and analyzing the historical training data and audio element information, the algorithm determines the cadence data of the exercise data as the third intensity cadence, so as to further meet the needs of users.
  • the audio element based method is realized.
  • the embodiment of the present application can further improve the first user's satisfaction with the generated motion data.
  • FIG. 4 shows a schematic diagram of an actual generation process of motion data provided by an exemplary embodiment of the present application.
  • the to-be-processed audio 410 corresponding to the first user is first input to the audio splitting model 420 , so that the audio splitting model 420 outputs audio layer information 430 .
  • the audio layer information 430 refers to the information obtained after the audio feature analysis and processing of the audio to be processed 410. Based on the audio layer information 430, the rhythm information 431, the beat information 432 and the intensity information 433 can be determined respectively.
  • FIG. 5 is a schematic flowchart of a method for assisting bicycle training provided by another exemplary embodiment of the present application.
  • the embodiment shown in FIG. 5 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 5 and the embodiment shown in FIG. 2 , and the similarities will not be repeated. .
  • Step S510 creating a virtual room for competition.
  • the virtual room for competition mentioned in step S510 is a virtual room established by the server and capable of presenting the competition information of the first user and the second user, and the virtual room can be displayed on the display screen of the user terminal.
  • the homeowner of the virtual room is the first user.
  • step S520 the invitation information of the first user is acquired, and the invitation information is sent to the corresponding second user.
  • the invitation information of the first user is acquired based on the battle virtual room.
  • the invitation information includes competition invitation information and/or companion invitation information.
  • the second user is also a user who wants to perform training based on a bicycle, and accordingly, the invitation information is sent to the user terminal of the corresponding second user.
  • Step S530 after receiving the confirmation and acceptance of the invitation information from the second user, establish a battle relationship between the first user and the second user, and send the to-be-processed audio and motion data to the second user.
  • first determine the to-be-processed audio corresponding to the first user input the to-be-processed audio into the audio splitting model to generate the audio element information corresponding to the to-be-processed audio, and generate the to-be-processed audio corresponding to the audio element information based on the audio element information sports data, then create a virtual room for battle, obtain the invitation information of the first user, and send the invitation information to the corresponding second user, and then after receiving the confirmation of the second user to accept the invitation information, establish the first user and the second user.
  • the battle relationship between users, and the pending audio and motion data are sent to the second user.
  • the method for assisting bicycle training provided by the embodiments of the present application can further meet the rich personalized training needs of the first user, meet the purpose of the first user socializing while training, and further improve the user experience favorability.
  • FIG. 6 is a schematic flowchart of a method for assisting bicycle training provided by yet another exemplary embodiment of the present application.
  • the embodiment shown in FIG. 6 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 6 and the embodiment shown in FIG. 5 , and the similarities will not be repeated. .
  • Step S610 during the battle, record the sports performance data of the first user and the second user.
  • the athletic performance data includes the degree of matching between the actual stepping rhythms of the first user and the second user and the rhythm information corresponding to the audio to be processed, and the like.
  • Step S620 performing a visual display operation on the sports performance data.
  • the sports performance data is visually displayed in the form of a Graphical User Interface (GUI).
  • GUI Graphical User Interface
  • the user terminal is set on the bicycle, and obtains sports performance data from the sensor set on the bicycle through Bluetooth.
  • GUI includes one or more of text, graphics, animation, and sound effects to present content in combination.
  • first determine the to-be-processed audio corresponding to the first user input the to-be-processed audio into the audio splitting model to generate the audio element information corresponding to the to-be-processed audio, and generate the to-be-processed audio corresponding to the audio element information based on the audio element information sports data, then create a virtual room for battle, obtain the invitation information of the first user, and send the invitation information to the corresponding second user, and then after receiving the confirmation of the second user to accept the invitation information, establish the first user and the second user.
  • the battle relationship between users, and the to-be-processed audio and sports data are sent to the second user.
  • the sports performance data of the first user and the second user is recorded, and the sports performance data is visualized.
  • the method for assisting bicycle training provided by the embodiments of the present application can further improve the interestingness of the training, thereby further improving the user experience favorability.
  • FIG. 7 shows a schematic flowchart of determining the audio to be processed corresponding to the first user according to an exemplary embodiment of the present application.
  • the embodiment shown in FIG. 7 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 7 and the embodiment shown in FIG. 2 , and the similarities will not be repeated. .
  • the step of determining the audio to be processed corresponding to the first user includes the following steps.
  • Step S211 acquiring audio voiceprint information input by the first user based on the audio input device.
  • the audio voiceprint information input by the first user may be audio voiceprint information sent by the first user humming or using a device such as an audio device.
  • the audio input device mentioned in step S211 can be either an audio input device mounted on a bicycle and connected to the server in communication, or an audio input device of a user terminal, such as a microphone of the user terminal.
  • Step S212 Determine index information corresponding to the audio voiceprint information based on the audio voiceprint information and a preset audio library.
  • the audio voiceprint information input by the first user is a segment of a song, not the entire song.
  • the song name information (ie index information) corresponding to the audio voiceprint information can be determined by comparing the audio voiceprint information with the songs in the preset audio library.
  • Step S213 determining the audio to be processed based on the index information.
  • the audio voiceprint information input by the first user is a segment of a song, and correspondingly, the audio to be processed is the complete audio of the song.
  • the audio voiceprint information input by the first user is acquired based on the audio input device, the index information corresponding to the audio voiceprint information is determined based on the audio voiceprint information and the preset audio library, and then the index information corresponding to the audio voiceprint information is determined based on the audio voiceprint information and the preset audio library.
  • the way in which the index information determines the audio to be processed achieves the purpose of determining the audio to be processed corresponding to the first user.
  • the embodiment of the present application does not require the first user to input the complete audio to be processed. Therefore, the embodiment of the present application can avoid that the first user cannot input the complete audio to be processed according to his own. The situation of training with interest greatly improves the user experience favorability.
  • FIG. 8 is a schematic flowchart of a training method for a network model provided by an exemplary embodiment of the present application. As shown in FIG. 8 , the training method of the network model provided by the embodiment of the present application includes the following steps.
  • Step S810 determining the training audio and the audio element information corresponding to the training audio.
  • the training audio mentioned in step S810 corresponds to the to-be-processed audio mentioned in the above embodiment.
  • both the training audio and the to-be-processed audio are audios corresponding to complete songs.
  • Step S820 an initial network model is established, and the initial network model is trained based on the training audio and audio element information to generate an audio splitting model.
  • the audio splitting model mentioned in step S820 is used to generate audio element information corresponding to the to-be-processed audio based on the to-be-processed audio.
  • the training method of the network model provided by the embodiment of the present application, by determining the training audio and the audio element information corresponding to the training audio, establishing an initial network model, and training the initial network model based on the training audio and the audio element information.
  • the purpose of splitting the model by determining the training audio and the audio element information corresponding to the training audio, establishing an initial network model, and training the initial network model based on the training audio and the audio element information.
  • FIG. 9 is a schematic structural diagram of an apparatus for assisting bicycle training provided by an exemplary embodiment of the present application.
  • the device for assisting bicycle training provided by the embodiment of the present application includes:
  • a to-be-processed audio determination module 910 configured to determine the to-be-processed audio corresponding to the first user
  • the first generation module 920 is used to input the audio to be processed into the audio splitting model to generate audio element information corresponding to the audio to be processed;
  • the second generating module 930 is configured to generate motion data corresponding to the audio to be processed based on the audio element information.
  • the device for assisting cycling training is a server.
  • FIG. 10 is a schematic structural diagram of a second generation module provided by an exemplary embodiment of the present application.
  • the embodiment shown in FIG. 10 of the present application is extended. The difference between the embodiment shown in FIG. 10 and the embodiment shown in FIG. 9 will be described below, and the similarities will not be repeated. .
  • the second generation module 930 includes:
  • a historical training data determining unit 931 configured to determine historical training data corresponding to the first user
  • the motion data generation unit 932 is configured to generate motion data based on historical training data and audio element information by using a preset data generation algorithm.
  • FIG. 11 is a schematic structural diagram of an apparatus for assisting bicycle training provided by another exemplary embodiment of the present application.
  • the embodiment shown in FIG. 11 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 11 and the embodiment shown in FIG. 9 , and the similarities will not be repeated. .
  • the device for assisting bicycle training provided by the embodiment of the present application further includes:
  • a creation module 1110 is used to create a virtual room for battle
  • the invitation information acquisition and sending module 1120 is used to obtain the invitation information of the first user, and send the invitation information to the corresponding second user;
  • the battle relationship establishing module 1130 is configured to establish a battle relationship between the first user and the second user after receiving the confirmation and acceptance of the invitation information from the second user, and send the to-be-processed audio and motion data to the second user.
  • FIG. 12 is a schematic structural diagram of an apparatus for assisting bicycle training provided by yet another exemplary embodiment of the present application.
  • the embodiment shown in FIG. 12 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 12 and the embodiment shown in FIG. 11 , and the similarities will not be repeated. .
  • the device for assisting bicycle training provided by the embodiment of the present application further includes:
  • the recording module 1210 is used to record the athletic performance data of the first user and the second user during the battle;
  • the presentation module 1220 is configured to perform a visual presentation operation on the sports performance data.
  • FIG. 13 is a schematic structural diagram of a to-be-processed audio determination module provided by an exemplary embodiment of the present application.
  • the embodiment shown in FIG. 13 of the present application is extended. The following focuses on the differences between the embodiment shown in FIG. 13 and the embodiment shown in FIG. 9 , and the similarities will not be repeated. .
  • the to-be-processed audio determination module 910 includes:
  • an audio voiceprint information obtaining unit 1310 configured to obtain the audio voiceprint information input by the first user based on the audio input device
  • an index information determining unit 1320 configured to determine index information corresponding to the audio voiceprint information based on the audio voiceprint information and a preset audio library;
  • the to-be-processed audio determination unit 1330 is configured to determine the to-be-processed audio based on the index information.
  • FIG. 14 is a schematic structural diagram of an apparatus for training a network model according to an exemplary embodiment of the present application.
  • the training device of the network model provided by the embodiment of the present application includes:
  • a determination module 1410 configured to determine training audio and audio element information corresponding to the training audio
  • the training module 1420 is used for establishing an initial network model, and training the initial network model based on the training audio and audio element information to generate an audio splitting model.
  • the training device of the network model is a server.
  • the operations and functions of the related modules and units mentioned in the apparatus for assisting bicycle training and the training apparatus for network models provided in FIGS. 9 to 14 can refer to the methods for assisting bicycle training and the training of network models provided in FIGS. 2 to 8 above.
  • the method, in order to avoid repetition, is not repeated here.
  • FIG. 15 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present application.
  • the electronic device 1500 includes one or more processors 1501 and a memory 1502 .
  • Processor 1501 may be a central processing unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in electronic device 1500 to perform desired functions.
  • CPU central processing unit
  • Processor 1501 may be a central processing unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in electronic device 1500 to perform desired functions.
  • Memory 1502 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory.
  • the volatile memory may include, for example, random access memory (RAM) and/or cache memory, or the like.
  • the non-volatile memory may include, for example, read only memory (ROM), hard disk, flash memory, and the like.
  • One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 1501 may execute the program instructions to implement the method and network for assisting bicycle training according to the various embodiments of the present application described above. The training method of the model and/or other desired features.
  • Various contents such as audio to be processed may also be stored in the computer-readable storage medium.
  • the electronic device 1500 may also include an input device 1503 and an output device 1504 interconnected by a bus system and/or other form of connection mechanism (not shown).
  • the input device 1503 may include, for example, a keyboard, a mouse, and the like.
  • the output device 1504 can output various information to the outside, including the determined motion data and the like.
  • the output device 1504 may include, for example, a display, a communication network and its connected remote output devices, and the like.
  • the electronic device 1500 may also include any other suitable components according to the specific application.
  • embodiments of the present application may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the "exemplary methods" described above in this specification.
  • the computer program product can write program codes for performing the operations of the embodiments of the present application in any combination of one or more programming languages, including object-oriented programming languages, such as Java, C++, etc. , also includes conventional procedural programming languages, such as "C" language or similar programming languages.
  • the program code may execute entirely on the user computing device, partly on the user device, as a stand-alone software package, partly on the user computing device and partly on a remote computing device, or entirely on the remote computing device or server execute on.
  • embodiments of the present application may also be computer-readable storage media having computer program instructions stored thereon, the computer program instructions, when executed by a processor, cause the processor to perform the above-mentioned "Example Method" section of this specification The steps in the method for assisting bicycle training and/or the method for training a network model according to various embodiments of the present application described in .
  • the computer-readable storage medium may employ any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses or devices, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
  • each component or each step can be decomposed and/or recombined. These disaggregations and/or recombinations should be considered as equivalents of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Computational Linguistics (AREA)
  • Cardiology (AREA)
  • Vascular Medicine (AREA)
  • General Health & Medical Sciences (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Electrically Operated Instructional Devices (AREA)

Abstract

一种辅助单车训练的方法及装置、网络模型的训练方法及装置，涉及信号处理技术领域。该辅助单车训练的方法包括：确定第一用户对应的待处理音频；将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息；基于音频元素信息生成待处理音频对应的运动数据。所述方法及装置能够基于第一用户对应的待处理音频生成用于辅助第一用户进行单车训练的运动数据，进而满足第一用户的个性化训练需求，提高用户体验好感度。此外，由于运动数据是基于音频拆分模型生成的音频元素信息确定的，运动数据和待处理音频之间的匹配度更高，能够进一步提升训练效果。

Description

辅助单车训练的方法及装置、网络模型的训练方法及装置 技术领域
本申请涉及信号处理技术领域，具体涉及辅助单车训练的方法及装置、网络模型的训练方法及装置、计算机可读存储介质和电子设备。
发明背景
近年来，随着经济的迅速发展，人们的生活质量在不断提高，人们的健身意识也在不断增强。动感单车作为一种有氧健身方式，不仅能够充分激活身体的运动细胞以消耗能量，而且能够实现减脂的目的。
然而，通常情况下，现有动感单车中的训练课程均是由教练预先录制生成的。具体而言，任一训练课程，不论是课程模式还是课程音乐，均是教练预先设定的。因此，现有动感单车的训练课程无法满足用户的个性化训练需求，用户体验好感度极差。
发明内容
为了解决上述技术问题，提出了本申请。本申请的实施例提供了一种辅助单车训练的方法及装置、网络模型的训练方法及装置、计算机可读存储介质和电子设备。
在一方面，本申请实施例提供了一种辅助单车训练的方法，该辅助单车训练的方法包括：确定第一用户对应的待处理音频；将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息；基于音频元素信息生成待处理音频对应的运动数据，其中，运动数据为用于辅助第一用户进行单车训练的运动数据。
在另一方面，本申请实施例提供了一种网络模型的训练方法，该网络模型的训练方法包括：确定训练音频以及训练音频对应的音频元素信息；建立初始网络模型，并基于训练音频和音频元素信息训练初始网络模型，以生成音频拆分模型，其中，音频拆分模型用于基于待处理音频生成待处理音频对应的音频元素信息。
在另一方面，本申请实施例提供了一种辅助单车训练的装置，该辅助单车训练的装置包括：待处理音频确定模块，用于确定第一用户对应的待处理音频；第一生成模块，用于将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息；第二生成模块，用于基于音频元素信息生成待处理音频对应的运动数据，其中，运动数据为用于辅助第一用户进行单车训练的运动数据。
在另一方面，本申请实施例提供了一种网络模型的训练装置，该网络模型的 训练装置包括：确定模块，用于确定训练音频以及训练音频对应的音频元素信息；训练模块，用于建立初始网络模型，并基于训练音频和音频元素信息训练初始网络模型，以生成音频拆分模型，其中，音频拆分模型用于基于待处理音频生成待处理音频对应的音频元素信息。
在另一方面，本申请实施例提供了一种计算机可读存储介质，所述存储介质存储有计算机程序，所述计算机程序用于执行上述实施例所述的辅助单车训练的方法，或者用于执行上述实施例所述的网络模型的训练方法。
在另一方面，本申请实施例提供了一种电子设备，所述电子设备包括：处理器；用于存储所述处理器可执行指令的存储器；所述处理器，用于执行上述实施例所述的辅助单车训练的方法，或者用于执行上述实施例所述的网络模型的训练方法。
在另一方面，本申请实施例提供了一种单车，该单车上装载有上述实施例所述的辅助单车训练的装置和/或网络模型的训练装置。
与现有技术相比，本申请实施例无需预先生成训练课程，并在训练课程中限定运动数据和运动数据对应的、能够辅助训练的音频。本申请实施例提供的辅助单车训练的方法，能够基于第一用户对应的待处理音频生成用于辅助第一用户进行单车训练的运动数据，进而满足第一用户的个性化训练需求，提高用户体验好感度。此外，由于运动数据是基于音频拆分模型生成的音频元素信息确定的，运动数据和待处理音频之间的匹配度更高，因此，本申请实施例也能够进一步提升训练效果。
附图简要说明
通过结合附图对本申请实施例进行更详细的描述，本申请的上述以及其他目的、特征和优势将变得更加明显。附图用来提供对本申请实施例的进一步理解，并且构成说明书的一部分，与本申请实施例一起用于解释本申请，并不构成对本申请的限制。在附图中，相同的参考标号通常代表相同部件或步骤。
图1所示为本申请实施例所适用的一场景示意图。
图2所示为本申请一示例性实施例提供的辅助单车训练的方法的流程示意图。
图3所示为本申请一示例性实施例提供的基于音频元素信息生成待处理音频对应的运动数据的流程示意图。
图4所示为本申请一示例性实施例提供的运动数据实际生成过程示意图。
图5所示为本申请另一示例性实施例提供的辅助单车训练的方法的流程示意图。
图6所示为本申请又一示例性实施例提供的辅助单车训练的方法的流程示意 图。
图7所示为本申请一示例性实施例提供的确定第一用户对应的待处理音频的流程示意图。
图8所示为本申请一示例性实施例提供的网络模型的训练方法的流程示意图。
图9所示为本申请一示例性实施例提供的辅助单车训练的装置的结构示意图。
图10所示为本申请一示例性实施例提供的第二生成模块的结构示意图。
图11所示为本申请另一示例性实施例提供的辅助单车训练的装置的结构示意图。
图12所示为本申请又一示例性实施例提供的辅助单车训练的装置的结构示意图。
图13所示为本申请一示例性实施例提供的待处理音频确定模块的结构示意图。
图14所示为本申请一示例性实施例提供的网络模型的训练装置的结构示意图。
图15所示为本申请一示例性实施例提供的电子设备的结构示意图。
实施本发明的方式
下面，将参考附图详细地描述根据本申请的示例实施例。显然，所描述的实施例仅仅是本申请的一部分实施例，而不是本申请的全部实施例，应理解，本申请不受这里描述的示例实施例的限制。
本申请实施例中的单车可以是户外骑行所用的普通单车，也可以是室内自行车训练课程所用的健身器材，本申请实施例对此不作限定。此外，本申请实施例中的用户终端可以是设置在单车上的用户终端，也可以是移动终端，例如手机或平板电脑。
示例性系统
图1所示为本申请实施例所适用的一场景示意图。如图1所示，本申请实施例所适用的场景中包括单车110和服务器120，其中，单车110上装载有用户终端111，服务器120与用户终端111之间具备通信连接关系。用户终端111用于获取第一用户的相关信息，并基于获取的相关信息与服务器120实现信息交互。服务器120中存储有音频拆分模型等数据。
具体而言，首先，服务器120基于单车110中的用户终端111确定第一用户对应的待处理音频，然后，服务器120将获取的待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息，基于音频元素信息生成待处理音频对应的运动数据，并将运动数据传送至用户终端111，以辅助第一用户进行单车训练。 即，该场景实现了一种辅助单车训练的方法。
在本申请实施例所适用的另一场景中，单车110进一步包括与用户终端111通信连接的传感器。传感器用于获取第一用户的运动表现数据，以便用户终端111或服务器120基于获取的运动表现数据进行运动评价等操作。举例说明，传感器设置在单车110的踏板内，进而借助设置在踏板内的传感器采集第一用户的踩踏力度、踩踏频率以及踩踏时间点等诸多运动信息。
示例性地，传感器和用户终端111基于蓝牙技术建立通信连接关系。
示例性方法
图2所示为本申请一示例性实施例提供的辅助单车训练的方法的流程示意图。如图2所示，本申请实施例提供的辅助单车训练的方法包括如下步骤。
步骤S210，确定第一用户对应的待处理音频。
示例性地，第一用户为想要借助单车进行单车训练的用户。
示例性地，待处理音频指的是与第一用户输入的音频声纹信息对应的待处理音频。
步骤S220，将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息。
在本申请一实施例中，音频元素信息包括节奏信息、节拍信息和能量信息中的至少一种。
示例性地，音频拆分模型为基于深度学习的神经网络模型，比如包括卷积层等结构的卷积神经网络模型。
步骤S230，基于音频元素信息生成待处理音频对应的运动数据。
步骤S230中提及的运动数据为用于辅助第一用户进行单车训练的运动数据。示例性地，运动数据包括踏频数据、速度数据和节奏数据中的至少一种。优选地，运动数据进一步包括分值数据、难度评级数据、最高分数据和分段分数数据中的至少一种。
由于音频元素信息能够更好地表征待处理音频的音频特性，因此，基于音频元素信息能够更准确地生成待处理音频对应的运动数据，从而为待处理音频匹配更适合的运动数据。示例性地，音频特性包括音频风格、音频类型和音频高潮区域等信息。
在实际应用过程中，首先确定第一用户对应的待处理音频，然后将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息，继而基于音频元素信息生成待处理音频对应的运动数据。
与现有技术相比，本申请实施例无需预先生成训练课程，并在训练课程中限定运动数据和运动数据对应的、能够辅助训练的音频。本申请实施例提供的辅助单车训练的方法，能够基于第一用户对应的待处理音频生成用于辅助第一用户进 行单车训练的运动数据，进而满足第一用户的个性化训练需求，提高用户体验好感度。此外，由于运动数据是基于音频拆分模型生成的音频元素信息确定的，运动数据和待处理音频之间的匹配度更高，因此，本申请实施例也能够进一步提升训练效果。
图3所示为本申请一示例性实施例提供的基于音频元素信息生成待处理音频对应的运动数据的流程示意图。在本申请图2所示实施例的基础上延伸出本申请图3所示实施例，下面着重叙述图3所示实施例与图2所示实施例的不同之处，相同之处不再赘述。
如图3所示，在本申请实施例提供的辅助单车训练的方法中，基于音频元素信息生成待处理音频对应的运动数据步骤，包括如下步骤。
步骤S231，确定第一用户对应的历史训练数据。
历史训练数据能够表征第一用户的运动能力和运动喜好等信息。那么，将历史训练数据作为生成运动数据的参考参数之一，能够进一步提高第一用户对所生成的运动数据的满意度。
示例性地，历史训练数据包括历史课程分数信息、历史课程匹配曲线信息、历史课程参与时长信息和历史训练时间信息中的至少一种。
步骤S232，利用预设数据生成算法，基于历史训练数据和音频元素信息生成运动数据。
需要说明的是，步骤S232中提及的预设数据生成算法，指的是能够综合历史训练数据和音频元素信息生成运动数据的算法。
举例说明，音频元素信息包括待处理音频的节拍信息，运动数据包括踏频数据，踏频数据具体包括第一强度踏频、第二强度踏频和第三强度踏频。如果基于节拍信息应当确定运动数据的踏频数据为第二强度踏频，而根据第一用户的历史训练数据发现第一用户的历史训练课程均是第三强度踏频，那么，预设数据生成算法通过对历史训练数据和音频元素信息的处理分析后，便将运动数据的踏频数据确定为第三强度踏频，以便进一步满足用户需求。
本申请实施例提供的辅助单车训练的方法，通过确定第一用户对应的历史训练数据，继而利用预设数据生成算法，基于历史训练数据和音频元素信息生成运动数据的方式，实现了基于音频元素信息生成待处理音频对应的运动数据的目的。本申请实施例能够进一步提高第一用户对所生成的运动数据的满意度。
下面结合图4具体说明运动数据的实际生成过程。
图4所示为本申请一示例性实施例提供的运动数据实际生成过程示意图。如图4所示，首先将第一用户对应的待处理音频410输入到音频拆分模型420，以便音频拆分模型420输出音频分层信息430。其中，音频分层信息430指的是对待处理音频410进行音频特征分析处理后得到的信息，基于音频分层信息430能 够分别确定节奏信息431、节拍信息432和强度信息433。
继续参照图4，利用预设数据生成算法450，基于节奏信息431、节拍信息432、强度信息433和第一用户的历史训练数据440生成能够辅助第一用户进行单车训练的运动数据460。
图5所示为本申请另一示例性实施例提供的辅助单车训练的方法的流程示意图。在本申请图2所示实施例的基础上延伸出本申请图5所示实施例，下面着重叙述图5所示实施例与图2所示实施例的不同之处，相同之处不再赘述。
如图5所示，在本申请实施例提供的辅助单车训练的方法中，在基于音频元素信息生成待处理音频对应的运动数据步骤之后，进一步包括如下步骤。
步骤S510，创建对战虚拟房间。
示例性地，步骤S510中提及的对战虚拟房间是服务器建立的、能够呈现第一用户和第二用户的竞赛信息的虚拟房间，该虚拟房间可以在用户终端的显示屏幕上显示。对战虚拟房间的房主为第一用户。
步骤S520，获取第一用户的邀请信息，并将邀请信息发送至对应的第二用户。
示例性地，基于对战虚拟房间获取第一用户的邀请信息。其中，邀请信息包括竞赛邀请信息和/或陪伴邀请信息。
优选地，第二用户亦为欲基于单车进行训练的用户，对应地，将邀请信息发送至对应的第二用户的用户终端。
步骤S530，接收到第二用户的确认接受邀请信息后，建立第一用户和第二用户之间的对战关系，并将待处理音频和运动数据发送至第二用户。
在实际应用过程中，首先确定第一用户对应的待处理音频，将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息，并基于音频元素信息生成待处理音频对应的运动数据，然后创建对战虚拟房间，获取第一用户的邀请信息，并将邀请信息发送至对应的第二用户，继而在接收到第二用户的确认接受邀请信息后，建立第一用户和第二用户之间的对战关系，并将待处理音频和运动数据发送至第二用户。
本申请实施例提供的辅助单车训练的方法，能够进一步满足第一用户丰富的个性化训练需求，满足第一用户在训练的同时进行社交的目的，进而进一步提高用户体验好感度。
图6所示为本申请又一示例性实施例提供的辅助单车训练的方法的流程示意图。在本申请图5所示实施例的基础上延伸出本申请图6所示实施例，下面着重叙述图6所示实施例与图5所示实施例的不同之处，相同之处不再赘述。
如图6所示，在本申请实施例提供的辅助单车训练的方法中，在接收到第二用户的确认接受邀请信息后，建立第一用户和第二用户之间的对战关系，并将待处理音频和运动数据发送至第二用户步骤之后，进一步包括如下步骤。
步骤S610，在对战过程中，记录第一用户和第二用户的运动表现数据。
示例性地，运动表现数据包括第一用户和第二用户的实际踩踏节奏与待处理音频对应的节奏信息的匹配度等。
步骤S620，将运动表现数据进行可视化展现操作。
在本申请一实施例中，借助用户终端，将运动表现数据以图形用户界面(Graphical User Interface，GUI)的形式进行可视化展示。其中，用户终端设置在单车上，并通过蓝牙从设置在单车上的传感器获取运动表现数据。
示例性地，图形用户界面包括文字、图表、动画以及音效等一种或多种的组合呈现内容。
在实际应用过程中，首先确定第一用户对应的待处理音频，将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息，并基于音频元素信息生成待处理音频对应的运动数据，然后创建对战虚拟房间，获取第一用户的邀请信息，并将邀请信息发送至对应的第二用户，继而在接收到第二用户的确认接受邀请信息后，建立第一用户和第二用户之间的对战关系，并将待处理音频和运动数据发送至第二用户，最后在对战过程中，记录第一用户和第二用户的运动表现数据，并将运动表现数据进行可视化展现操作。
本申请实施例提供的辅助单车训练的方法，能够进一步提升训练的趣味性，进而进一步提高用户体验好感度。
图7所示为本申请一示例性实施例提供的确定第一用户对应的待处理音频的流程示意图。在本申请图2所示实施例的基础上延伸出本申请图7所示实施例，下面着重叙述图7所示实施例与图2所示实施例的不同之处，相同之处不再赘述。
如图7所示，在本申请实施例提供的辅助单车训练的方法中，确定第一用户对应的待处理音频步骤，包括如下步骤。
步骤S211，基于音频输入装置获取第一用户输入的音频声纹信息。
在本申请一实施例中，第一用户输入的音频声纹信息可以是第一用户自行哼唱或借助音响等设备发出的音频声纹信息。
步骤S211中提及的音频输入装置，既可以是单车上装载的、与服务器通信连接的音频输入装置，又可以是用户终端的音频输入装置，比如用户终端的麦克风。
步骤S212，基于音频声纹信息和预设音频库确定音频声纹信息对应的索引信息。
示例性地，第一用户输入的音频声纹信息是一首歌曲的一个片段，并非整首歌曲。那么，本申请实施例通过将音频声纹信息与预设音频库中的歌曲进行比对的方式能够确定该音频声纹信息对应的歌曲名称信息(即索引信息)。
步骤S213，基于索引信息确定待处理音频。
示例性地，第一用户输入的音频声纹信息是一首歌曲的一个片段，对应地， 待处理音频为该首歌曲的完整音频。
本申请实施例提供的辅助单车训练的方法，通过基于音频输入装置获取第一用户输入的音频声纹信息，基于音频声纹信息和预设音频库确定音频声纹信息对应的索引信息，继而基于索引信息确定待处理音频的方式，实现了确定第一用户对应的待处理音频的目的。与现有技术相比，本申请实施例无需第一用户输入完整的待处理音频，因此，本申请实施例能够避免因第一用户不能输入完整的待处理音频，第一用户便不能依照自己的兴趣进行训练的情况，极大提高了用户体验好感度。
图8所示为本申请一示例性实施例提供的网络模型的训练方法的流程示意图。如图8所示，本申请实施例提供的网络模型的训练方法包括如下步骤。
步骤S810，确定训练音频以及训练音频对应的音频元素信息。
步骤S810中提及的训练音频与上述实施例提及的待处理音频对应。比如，训练音频和待处理音频均为完整歌曲对应的音频。
步骤S820，建立初始网络模型，并基于训练音频和音频元素信息训练初始网络模型，以生成音频拆分模型。
步骤S820中提及的音频拆分模型用于基于待处理音频生成待处理音频对应的音频元素信息。
本申请实施例提供的网络模型的训练方法，通过确定训练音频以及训练音频对应的音频元素信息，建立初始网络模型，并基于训练音频和音频元素信息训练初始网络模型的方式，实现了训练生成音频拆分模型的目的。
示例性装置
图9所示为本申请一示例性实施例提供的辅助单车训练的装置的结构示意图。如图9所示，本申请实施例提供的辅助单车训练的装置包括：
待处理音频确定模块910，用于确定第一用户对应的待处理音频；
第一生成模块920，用于将待处理音频输入至音频拆分模型，以生成待处理音频对应的音频元素信息；
第二生成模块930，用于基于音频元素信息生成待处理音频对应的运动数据。
示例性地，辅助单车训练的装置为服务器。
图10所示为本申请一示例性实施例提供的第二生成模块的结构示意图。在本申请图9所示实施例的基础上延伸出本申请图10所示实施例，下面着重叙述图10所示实施例与图9所示实施例的不同之处，相同之处不再赘述。
如图10所示，在本申请实施例提供的辅助单车训练的装置中，第二生成模块930包括：
历史训练数据确定单元931，用于确定第一用户对应的历史训练数据；
运动数据生成单元932，用于利用预设数据生成算法，基于历史训练数据和 音频元素信息生成运动数据。
图11所示为本申请另一示例性实施例提供的辅助单车训练的装置的结构示意图。在本申请图9所示实施例的基础上延伸出本申请图11所示实施例，下面着重叙述图11所示实施例与图9所示实施例的不同之处，相同之处不再赘述。
如图11所示，本申请实施例提供的辅助单车训练的装置进一步包括：
创建模块1110，用于创建对战虚拟房间；
邀请信息获取并发送模块1120，用于获取第一用户的邀请信息，并将邀请信息发送至对应的第二用户；
对战关系建立模块1130，用于接收到第二用户的确认接受邀请信息后，建立第一用户和第二用户之间的对战关系，并将待处理音频和运动数据发送至第二用户。
图12所示为本申请又一示例性实施例提供的辅助单车训练的装置的结构示意图。在本申请图11所示实施例的基础上延伸出本申请图12所示实施例，下面着重叙述图12所示实施例与图11所示实施例的不同之处，相同之处不再赘述。
如图12所示，本申请实施例提供的辅助单车训练的装置进一步包括：
记录模块1210，用于在对战过程中，记录第一用户和第二用户的运动表现数据；
展现模块1220，用于将运动表现数据进行可视化展现操作。
图13所示为本申请一示例性实施例提供的待处理音频确定模块的结构示意图。在本申请图9所示实施例的基础上延伸出本申请图13所示实施例，下面着重叙述图13所示实施例与图9所示实施例的不同之处，相同之处不再赘述。
如图13所示，在本申请实施例提供的辅助单车训练的装置中，待处理音频确定模块910包括：
音频声纹信息获取单元1310，用于基于音频输入装置获取第一用户输入的音频声纹信息；
索引信息确定单元1320，用于基于音频声纹信息和预设音频库确定音频声纹信息对应的索引信息；
待处理音频确定单元1330，用于基于索引信息确定待处理音频。
图14所示为本申请一示例性实施例提供的网络模型的训练装置的结构示意图。如图14所示，本申请实施例提供的网络模型的训练装置包括：
确定模块1410，用于确定训练音频以及训练音频对应的音频元素信息；
训练模块1420，用于建立初始网络模型，并基于训练音频和音频元素信息训练初始网络模型，以生成音频拆分模型。
示例性地，网络模型的训练装置为服务器。
图9至图14提供的辅助单车训练的装置和网络模型的训练装置中提及的相关 模块以及单元的操作和功能可以参考上述图2至图8提供的辅助单车训练的方法和网络模型的训练方法，为了避免重复，在此不再赘述。
示例性电子设备
下面，参考图15来描述根据本申请实施例的电子设备。图15所示为本申请一示例性实施例提供的电子设备的结构示意图。
如图15所示，电子设备1500包括一个或多个处理器1501和存储器1502。
处理器1501可以是中央处理单元(CPU)或者具有数据处理能力和/或指令执行能力的其他形式的处理单元，并且可以控制电子设备1500中的其他组件以执行期望的功能。
存储器1502可以包括一个或多个计算机程序产品，所述计算机程序产品可以包括各种形式的计算机可读存储介质，例如易失性存储器和/或非易失性存储器。所述易失性存储器例如可以包括随机存取存储器(RAM)和/或高速缓冲存储器(cache)等。所述非易失性存储器例如可以包括只读存储器(ROM)、硬盘、闪存等。在所述计算机可读存储介质上可以存储一个或多个计算机程序指令，处理器1501可以运行所述程序指令，以实现上文所述的本申请的各个实施例的辅助单车训练的方法、网络模型的训练方法以及/或者其他期望的功能。在所述计算机可读存储介质中还可以存储诸如待处理音频等各种内容。
在一个示例中，电子设备1500还可以包括：输入装置1503和输出装置1504，这些组件通过总线系统和/或其他形式的连接机构(未示出)互连。
该输入装置1503可以包括例如键盘、鼠标等等。
该输出装置1504可以向外部输出各种信息，包括确定出的运动数据等。该输出装置1504可以包括例如显示器、通信网络及其所连接的远程输出设备等等。
当然，为了简化，图15中仅示出了该电子设备1500中与本申请有关的组件中的一些，省略了诸如总线、输入/输出接口等等的组件。除此之外，根据具体应用情况，电子设备1500还可以包括任何其他适当的组件。
示例性计算机程序产品和计算机可读存储介质
除了上述方法和设备以外，本申请的实施例还可以是计算机程序产品，其包括计算机程序指令，所述计算机程序指令在被处理器运行时使得所述处理器执行本说明书上述“示例性方法”部分中描述的根据本申请各种实施例的辅助单车训练的方法和/或网络模型的训练方法中的步骤。
所述计算机程序产品可以以一种或多种程序设计语言的任意组合来编写用于执行本申请实施例操作的程序代码，所述程序设计语言包括面向对象的程序设计语言，诸如Java、C++等，还包括常规的过程式程序设计语言，诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算设备上执行、部分地在用户设备上执行、作为一个独立的软件包执行、部分在用户计算设备上部分在远程 计算设备上执行、或者完全在远程计算设备或服务器上执行。
此外，本申请的实施例还可以是计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令在被处理器运行时使得所述处理器执行本说明书上述“示例性方法”部分中描述的根据本申请各种实施例的辅助单车训练的方法和/或网络模型的训练方法中的步骤。
所述计算机可读存储介质可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以包括但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
以上结合具体实施例描述了本申请的基本原理，但是，需要指出的是，在本申请中提及的优点、优势、效果等仅是示例而非限制，不能认为这些优点、优势、效果等是本申请的各个实施例必须具备的。另外，上述公开的具体细节仅是为了示例的作用和便于理解的作用，而非限制，上述细节并不限制本申请为必须采用上述具体的细节来实现。
本申请中涉及的器件、装置、设备、系统的方框图仅作为例示性的例子并且不意图要求或暗示必须按照方框图示出的方式进行连接、布置、配置。如本领域技术人员将认识到的，可以按任意方式连接、布置、配置这些器件、装置、设备、系统。诸如“包括”、“包含”、“具有”等等的词语是开放性词汇，指“包括但不限于”，且可与其互换使用。这里所使用的词汇“或”和“和”指词汇“和/或”，且可与其互换使用，除非上下文明确指示不是如此。这里所使用的词汇“诸如”指词组“诸如但不限于”，且可与其互换使用。
还需要指出的是，在本申请的装置、设备和方法中，各部件或各步骤是可以分解和/或重新组合的。这些分解和/或重新组合应视为本申请的等效方案。
提供所公开的方面的以上描述以使本领域的任何技术人员能够做出或者使用本申请。对这些方面的各种修改对于本领域技术人员而言是非常显而易见的，并且在此定义的一般原理可以应用于其他方面而不脱离本申请的范围。因此，本申请不意图被限制到在此示出的方面，而是按照与在此发明的原理和新颖的特征一致的最宽范围。
为了例示和描述的目的已经给出了以上描述。此外，此描述不意图将本申请的实施例限制到在此发明的形式。尽管以上已经讨论了多个示例方面和实施例，但是本领域技术人员将认识到其某些变型、修改、改变、添加和子组合。

Claims (19)

  1. 一种辅助单车训练的方法，包括：
    确定第一用户对应的待处理音频；
    将所述待处理音频输入至音频拆分模型，以生成所述待处理音频对应的音频元素信息；
    基于所述音频元素信息生成所述待处理音频对应的运动数据，其中，所述运动数据为用于辅助所述第一用户进行单车训练的运动数据。
  2. 根据权利要求1所述的辅助单车训练的方法，其中，所述基于所述音频元素信息生成所述待处理音频对应的运动数据，包括：
    确定所述第一用户对应的历史训练数据；
    利用预设数据生成算法，基于所述历史训练数据和所述音频元素信息生成所述运动数据。
  3. 根据权利要求2所述的辅助单车训练的方法，其中，所述历史训练数据包括历史课程分数信息、历史课程匹配曲线信息、历史课程参与时长信息和历史训练时间信息中的至少一种。
  4. 根据权利要求1至3任一项所述的辅助单车训练的方法，其中，在所述基于所述音频元素信息生成所述待处理音频对应的运动数据之后，进一步包括：
    创建对战虚拟房间，其中，所述对战虚拟房间的房主为所述第一用户；
    获取所述第一用户的邀请信息，并将所述邀请信息发送至对应的第二用户；
    接收到所述第二用户的确认接受邀请信息后，建立所述第一用户和所述第二用户之间的对战关系，并将所述待处理音频和所述运动数据发送至所述第二用户。
  5. 根据权利要求4所述的辅助单车训练的方法，其中，在所述建立所述第一用户和所述第二用户之间的对战关系，并将所述待处理音频和所述运动数据发送至所述第二用户之后，进一步包括：
    在对战过程中，记录所述第一用户和所述第二用户的运动表现数据；
    将所述运动表现数据进行可视化展现操作。
  6. 根据权利要求5所述的辅助单车训练的方法，其中，所述将所述运动表现数据进行可视化展现操作包括：
    将所述运动表现数据以图形用户界面的形式进行可视化展示。
  7. 根据权利要求6所述的辅助单车训练的方法，其中，所述图形用户界面包括文字、图表、动画以及音效中的一种或多种的组合呈现内容。
  8. 根据权利要求1至7任一项所述的辅助单车训练的方法，其中，所述确定第一用户对应的待处理音频，包括：
    基于音频输入装置获取所述第一用户输入的音频声纹信息；
    基于所述音频声纹信息和预设音频库确定所述音频声纹信息对应的索引信息；
    基于所述索引信息确定所述待处理音频。
  9. 根据权利要求8所述的辅助单车训练的方法，其中，所述第一用户输入的所述音频声纹信息包括所述第一用户自行哼唱或借助音响发出的音频声纹信息。
  10. 根据权利要求1至9任一项所述的辅助单车训练的方法，其中，所述音频元素信息包括节奏信息、节拍信息和能量信息中的至少一种。
  11. 根据权利要求1至10任一项所述的辅助单车训练的方法，其中，所述运动数据包括踏频数据、速度数据和节奏数据中的至少一种。
  12. 根据权利要求11所述的辅助单车训练的方法，其中，所述踏频数据包括：第一强度踏频、第二强度踏频和第三强度踏频。
  13. 根据权利要求1至12任一项所述的辅助单车训练的方法，其中，所述运动数据进一步包括分值数据、难度评级数据、最高分数据和分段分数数据中的至少一种。
  14. 一种网络模型的训练方法，包括：
    确定训练音频以及训练音频对应的音频元素信息；
    建立初始网络模型，并基于所述训练音频和所述音频元素信息训练所述初始网络模型，以生成音频拆分模型，其中，所述音频拆分模型用于基于待处理音频生成所述待处理音频对应的音频元素信息。
  15. 一种辅助单车训练的装置，包括：
    待处理音频确定模块，用于确定第一用户对应的待处理音频；
    第一生成模块，用于将所述待处理音频输入至音频拆分模型，以生成所述待处理音频对应的音频元素信息；
    第二生成模块，用于基于所述音频元素信息生成所述待处理音频对应的运动数据，其中，所述运动数据为用于辅助所述第一用户进行单车训练的运动数据。
  16. 一种网络模型的训练装置，包括：
    确定模块，用于确定训练音频以及训练音频对应的音频元素信息；
    训练模块，用于建立初始网络模型，并基于所述训练音频和所述音频元素信息训练所述初始网络模型，以生成音频拆分模型，其中，所述音频拆分模型用于基于待处理音频生成所述待处理音频对应的音频元素信息。
  17. 一种计算机可读存储介质，所述存储介质存储有计算机程序，所述计算机程序用于执行上述权利要求1至13任一项所述的辅助单车训练的方法，或者用于执行上述权利要求14所述的网络模型的训练方法。
  18. 一种电子设备，所述电子设备包括：
    处理器；
    用于存储所述处理器可执行指令的存储器；
    所述处理器，用于执行上述权利要求1至13任一项所述的辅助单车训练的方法，或者用于执行上述权利要求14所述的网络模型的训练方法。
  19. 一种单车，装载有如上述权利要求15所述的辅助单车训练的装置和/或如上述权利要求16所述的网络模型的训练装置。
PCT/CN2021/104029 2020-07-01 2021-07-01 辅助单车训练的方法及装置、网络模型的训练方法及装置 WO2022002204A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010620684.0 2020-07-01
CN202010620684.0A CN111773620A (zh) 2020-07-01 2020-07-01 辅助单车训练的方法及装置、网络模型的训练方法及装置

Publications (1)

Publication Number Publication Date
WO2022002204A1 true WO2022002204A1 (zh) 2022-01-06

Family

ID=72760942

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/104029 WO2022002204A1 (zh) 2020-07-01 2021-07-01 辅助单车训练的方法及装置、网络模型的训练方法及装置

Country Status (2)

Country Link
CN (1) CN111773620A (zh)
WO (1) WO2022002204A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112973092B (zh) * 2021-02-02 2022-03-25 随机漫步(上海)体育科技有限公司 辅助训练的方法及装置、存储介质、电子设备和单车
CN111773620A (zh) * 2020-07-01 2020-10-16 随机漫步(上海)体育科技有限公司 辅助单车训练的方法及装置、网络模型的训练方法及装置

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5489249A (en) * 1991-07-02 1996-02-06 Proform Fitness Products, Inc. Video exercise control system
US5645509A (en) * 1991-07-02 1997-07-08 Icon Health & Fitness, Inc. Remote exercise control system
CN1596142A (zh) * 2001-09-05 2005-03-16 艾肯Ip有限公司 使一个或多个锻炼设备与计算机之间能够双向通信和使一个或多个锻炼设备的用户能够进行竞争性锻炼的系统与方法
CN101454050A (zh) * 2006-03-22 2009-06-10 拉尔夫·J·德尔乔诺 虚拟个人训练设备
US20100259407A1 (en) * 2009-04-08 2010-10-14 Polar Electro Oy Portable Apparatus
CN103578062A (zh) * 2012-07-20 2014-02-12 北京知康优美科技有限公司 一种基于移动互联和二维码技术的科学健身指导实现方法与系统
CN104126288A (zh) * 2012-05-31 2014-10-29 冯林 一种对健身提供帮助的系统
CN105513583A (zh) * 2015-11-25 2016-04-20 福建星网视易信息系统有限公司 一种歌曲节奏的显示方法及其系统
CN105892628A (zh) * 2015-08-11 2016-08-24 乐视体育文化产业发展（北京）有限公司 音乐推荐方法、装置和自行车
CN107349573A (zh) * 2016-05-10 2017-11-17 北京环形山网络科技有限公司 基于智能手机和物联网的虚拟现实健身系统
CN107737439A (zh) * 2017-10-18 2018-02-27 徐�明 一种进阶式体能训练系统及方法
CN107823846A (zh) * 2017-09-13 2018-03-23 华东师范大学 一种带有体质检测的共享式虚拟现实健身骑行系统及方法
CN107930085A (zh) * 2017-12-22 2018-04-20 武汉市龙五物联网络科技有限公司 一种带有娱乐交互系统的多功能健身设备
CN108635753A (zh) * 2018-05-28 2018-10-12 苏州创存数字科技有限公司 一种教学用动感单车的控制方法及动感单车
CN108853946A (zh) * 2018-07-10 2018-11-23 燕山大学 一种基于Kinect的健身指导训练系统及方法
CN109908551A (zh) * 2019-02-02 2019-06-21 陈骋 一种单车锻炼方法及装置
CN111773620A (zh) * 2020-07-01 2020-10-16 随机漫步(上海)体育科技有限公司 辅助单车训练的方法及装置、网络模型的训练方法及装置

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2547348Y (zh) * 2002-04-19 2003-04-30 黄加琛 利用音乐控制速度的电动跑步机
CN202590245U (zh) * 2012-05-10 2012-12-12 山东科技大学 多功能动感单车
US9269341B1 (en) * 2013-06-30 2016-02-23 Second Wind Technologies, Inc. Method for processing music to match runners tempo
CN105944332B (zh) * 2016-05-10 2018-12-14 杭州韵健科技有限公司 一种可联网的虚拟现实智能健身系统
CN110675886B (zh) * 2019-10-09 2023-09-15 腾讯科技（深圳）有限公司 音频信号处理方法、装置、电子设备及存储介质

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5489249A (en) * 1991-07-02 1996-02-06 Proform Fitness Products, Inc. Video exercise control system
US5645509A (en) * 1991-07-02 1997-07-08 Icon Health & Fitness, Inc. Remote exercise control system
CN1596142A (zh) * 2001-09-05 2005-03-16 艾肯Ip有限公司 使一个或多个锻炼设备与计算机之间能够双向通信和使一个或多个锻炼设备的用户能够进行竞争性锻炼的系统与方法
CN101454050A (zh) * 2006-03-22 2009-06-10 拉尔夫·J·德尔乔诺 虚拟个人训练设备
US20100259407A1 (en) * 2009-04-08 2010-10-14 Polar Electro Oy Portable Apparatus
CN104126288A (zh) * 2012-05-31 2014-10-29 冯林 一种对健身提供帮助的系统
CN103578062A (zh) * 2012-07-20 2014-02-12 北京知康优美科技有限公司 一种基于移动互联和二维码技术的科学健身指导实现方法与系统
CN105892628A (zh) * 2015-08-11 2016-08-24 乐视体育文化产业发展（北京）有限公司 音乐推荐方法、装置和自行车
CN105513583A (zh) * 2015-11-25 2016-04-20 福建星网视易信息系统有限公司 一种歌曲节奏的显示方法及其系统
CN107349573A (zh) * 2016-05-10 2017-11-17 北京环形山网络科技有限公司 基于智能手机和物联网的虚拟现实健身系统
CN107823846A (zh) * 2017-09-13 2018-03-23 华东师范大学 一种带有体质检测的共享式虚拟现实健身骑行系统及方法
CN107737439A (zh) * 2017-10-18 2018-02-27 徐�明 一种进阶式体能训练系统及方法
CN107930085A (zh) * 2017-12-22 2018-04-20 武汉市龙五物联网络科技有限公司 一种带有娱乐交互系统的多功能健身设备
CN108635753A (zh) * 2018-05-28 2018-10-12 苏州创存数字科技有限公司 一种教学用动感单车的控制方法及动感单车
CN108853946A (zh) * 2018-07-10 2018-11-23 燕山大学 一种基于Kinect的健身指导训练系统及方法
CN109908551A (zh) * 2019-02-02 2019-06-21 陈骋 一种单车锻炼方法及装置
CN111773620A (zh) * 2020-07-01 2020-10-16 随机漫步(上海)体育科技有限公司 辅助单车训练的方法及装置、网络模型的训练方法及装置

Also Published As

Publication number Publication date
CN111773620A (zh) 2020-10-16

Similar Documents

Publication Publication Date Title
WO2022002204A1 (zh) 辅助单车训练的方法及装置、网络模型的训练方法及装置
US11205408B2 (en) Method and system for musical communication
US20200126566A1 (en) Method and apparatus for voice interaction
CN104460982A (zh) 基于生物统计参数给出音频
US11511200B2 (en) Game playing method and system based on a multimedia file
CN110808038B (zh) 普通话评测方法、装置、设备及存储介质
WO2022184055A1 (zh) 文章的语音播放方法、装置、设备、存储介质及程序产品
JP7135063B2 (ja) スマート音声機器の制御方法、装置、電子機器及び記憶媒体
CN109741724B (zh) 制作歌曲的方法、装置及智能音响
CN112973092B (zh) 辅助训练的方法及装置、存储介质、电子设备和单车
US20150055934A1 (en) Enhancing karaoke systems utilizing audience sentiment feedback and audio watermarking
CN117541444B (zh) 一种互动虚拟现实口才表达训练方法、装置、设备及介质
US20240340506A1 (en) Method and apparatus for assisting watching video content
CN107770235A (zh) 一种斗歌服务实现方法及系统
WO2021176925A1 (ja) 演奏データに対する観衆の評価を推論する方法、システム、及びプログラム
WO2021193033A1 (ja) 訓練済みモデルの確立方法、推定方法、演奏エージェントの推薦方法、演奏エージェントの調整方法、訓練済みモデルの確立システム、推定システム、訓練済みモデルの確立プログラム及び推定プログラム
Lee et al. Improving orchestral conducting systems in public spaces: examining the temporal characteristics and conceptual models of conducting gestures
US9928830B2 (en) Information processing apparatus and information processing method
CN111450484B (zh) 辅助单车训练的方法、可读存储介质及电子设备
US20220020390A1 (en) Interactive group session computing systems and related methods
WO2021197444A1 (zh) 辅助单车训练的方法、服务器、用户终端及训练单车
CN111450483A (zh) 辅助单车训练的方法、可读存储介质和电子设备
Mishra et al. Music tune generation based on facial emotion
TW201426730A (zh) 音樂的情緒分類方法
TWI522954B (zh) 用以輔助健身的多媒體編輯方法

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: 21833514

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: 21833514

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 03.07.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21833514

Country of ref document: EP

Kind code of ref document: A1