WO2023273156A1 - 时延估计调整方法、装置、设备以及存储介质 - Google Patents

时延估计调整方法、装置、设备以及存储介质 Download PDF

Info

Publication number
WO2023273156A1
WO2023273156A1 PCT/CN2021/135071 CN2021135071W WO2023273156A1 WO 2023273156 A1 WO2023273156 A1 WO 2023273156A1 CN 2021135071 W CN2021135071 W CN 2021135071W WO 2023273156 A1 WO2023273156 A1 WO 2023273156A1
Authority
WO
WIPO (PCT)
Prior art keywords
delay
delay value
value
playback
maximum value
Prior art date
Application number
PCT/CN2021/135071
Other languages
English (en)
French (fr)
Inventor
刘嵘
Original Assignee
阿波罗智联(北京)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 阿波罗智联(北京)科技有限公司 filed Critical 阿波罗智联(北京)科技有限公司
Publication of WO2023273156A1 publication Critical patent/WO2023273156A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R1/00—Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
    • B60R1/02—Rear-view mirror arrangements
    • 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
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/02—Methods for producing synthetic speech; Speech synthesisers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0852—Delays

Definitions

  • the present disclosure relates to the field of artificial intelligence, in particular to the fields of computer technology and intelligent transportation.
  • the smart rearview mirror can transmit audio to the original car machine for playback through screen projection technology.
  • This kind of delay has a great negative impact on the voice signal processing, because of the limitation of the hardware noise reduction chip, the current signal processing technology cannot be performed in the case of unconventional (>40ms) delay. Therefore, the smart rearview mirror introduces a delay estimation algorithm, which calculates the delay and delays the voice signal.
  • the delay estimation algorithm usually limits the maximum value of the delay estimation to a smaller range (for example, 1000ms). However, because the performance of each native car that supports screen projection is inconsistent, the real transmission delay fluctuates greatly. If the maximum value of the time delay estimation is uniformly increased at this time, the problem of longer convergence time will appear again.
  • the disclosure provides a delay estimation adjustment method, device, equipment, storage medium and program product.
  • a delay estimation adjustment method including: initializing the maximum value of the delay estimate; estimating the first delay value during the playback of audition music; Estimating a second delay value; adjusting a maximum value of the delay estimate based on the first delay value and the second delay value.
  • an apparatus for adjusting delay estimation including: a setting module configured to initialize the maximum value of delay estimation; a first estimation module configured to estimate The first time delay value; the second estimation module is configured to estimate the second time delay value during the sound test synthesis speech playback process; the adjustment module is configured to adjust the time delay value based on the first time delay value and the second time delay value The maximum value of the delay estimate.
  • an electronic device including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by Executed by at least one processor, so that at least one processor can execute the method described in any implementation manner of the first aspect.
  • a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method described in any implementation manner of the first aspect.
  • a computer program product including a computer program, and when the computer program is executed by a processor, the method described in any implementation manner of the first aspect is implemented.
  • the disclosure can adjust the maximum value of time delay estimation in a targeted manner, and greatly shorten the convergence time of the algorithm.
  • FIG. 1 is a flowchart of an embodiment of a delay estimation adjustment method according to the present disclosure
  • FIG. 2 is a flow chart of another embodiment of a delay estimation adjustment method according to the present disclosure.
  • FIG. 3 is a flowchart of another embodiment of a delay estimation adjustment method according to the present disclosure.
  • FIG. 4 is a scene diagram that can implement a delay estimation adjustment method according to an embodiment of the present disclosure
  • FIG. 5 is a schematic structural diagram of an embodiment of a device for adjusting delay estimation according to the present disclosure
  • Fig. 6 is a block diagram of an electronic device for implementing the delay estimation adjustment method of the embodiment of the present disclosure.
  • FIG. 1 shows a flow 100 of an embodiment of a delay estimation adjustment method according to the present disclosure.
  • the delay estimation adjustment method includes the following steps:
  • Step 101 initialize the maximum value of time delay estimation.
  • the execution subject of the delay estimation adjusting method may initialize the maximum value of the delay estimation.
  • the estimated delay value will be limited by the maximum value of the delay estimate and will not exceed the maximum value of the delay estimate.
  • the speech interference signal can be processed thoroughly.
  • the speech interference signal cannot be completely processed.
  • the end user can initiate an automatic adjustment request for delay estimation, and the voice assistant will enter the corresponding mode after connection, and initialize the maximum value of delay estimation.
  • the maximum value of the delay estimation is usually initialized to a relatively large value.
  • the above-mentioned executive body may set the maximum value of the delay estimation (for example, 2500ms) as the upper limit value of the delay estimation algorithm. Since the delay value of any audio will not exceed the upper limit value of the delay estimation algorithm, the maximum value of the delay estimation is initially set as the upper limit value of the delay estimation algorithm, so that the speech interference signal of any audio frequency can be completely eliminated. deal with.
  • Step 102 estimating the first time delay value during the playing of the audition music.
  • the execution subject may estimate the first delay value during the playback of the audition music.
  • smart rearview mirrors can transmit music to the original car for playback through screen projection technology. Due to the huge amount of music data, it takes a certain amount of time to transmit from the smart rearview mirror to the original car, resulting in a certain delay in music playback. Therefore, it is necessary to estimate the delay value from the transmission of the test sound music by the smart rearview mirror to the playback of the test sound music by the original car during the playback of the test sound music, that is, the first delay value. Specifically, the smart rearview mirror transmits the audition music to the original car machine through the screen projection technology, and the original car machine starts to play the audition music. In the process of playing the audition music, the time delay value of the audition music is estimated by using the time delay estimation algorithm to obtain the first time delay value.
  • the time delay value of the audition music is estimated by using the time delay estimation algorithm as the first time delay value.
  • the audition music may be any music played for adjusting the maximum value of time delay estimation.
  • the audition music playback process is the process of transmitting the audition music from the smart rearview mirror to the original car machine playing the audition music.
  • the delay estimation algorithm can be used to estimate the delay value of audio playback, including but not limited to the currently existing delay estimation algorithm and the delay estimation algorithm that may appear in the future.
  • the ETDE algorithm is a time-delay estimation method with constraints.
  • the filter uses an N-order interpolation operation, and the instantaneous time-delay estimation is directly used instead of the filter weight coefficient for iteration, which reduces the amount of calculation.
  • Step 103 estimating a second time delay value during the playback of the sound test synthesized voice.
  • the execution subject may estimate the second delay value during the playback of the sound test synthesized voice.
  • the smart rearview mirror can transmit the synthesized voice to the original car machine for playback through screen projection technology. Due to the huge amount of synthesized voice data, it takes a certain amount of time to transmit from the smart rearview mirror to the original car, resulting in a certain delay in the playback of synthesized voice. Therefore, it is necessary to estimate the delay value from the transmission of the sound test synthesized voice by the smart rearview mirror to the play of the test sound synthesized voice by the original vehicle during the playback of the sound test synthesized voice, that is, the second delay value. Specifically, the smart rearview mirror transmits the synthesized voice of the test sound to the original car machine through the projection technology, and the original car machine starts to play the synthesized voice of the test sound.
  • the delay value of the synthesized voice of audition is estimated by using the time delay estimation algorithm, so as to obtain the second delay value.
  • the time delay value of the synthesized speech in a sound trial is estimated by using a time delay estimation algorithm as the second time delay value.
  • the synthesized voice may be a voice command generated by the smart rearview mirror based on collected information.
  • the sound test synthetic speech may be any synthetic speech played for adjusting the maximum value of the delay estimate.
  • the process of playing the sound test synthesized voice is the process of transmitting the test sound synthesized voice from the smart rearview mirror to the original car machine to play the test sound synthesized voice.
  • Step 104 based on the first delay value and the second delay value, adjust the maximum value of the delay estimate.
  • the execution subject may adjust the maximum value of the estimated delay based on the first delay value and the second delay value.
  • the maximum value of the estimated delay can be adjusted based on any one of the first delay value and the second delay value.
  • one of the first delay value or the second delay value is used as the maximum value of the delay estimate.
  • the first delay value or one of the delay values is increased by a preset duration as the maximum value of the delay estimate.
  • the maximum value of the adjusted delay estimation is smaller than the upper limit value of the delay estimation algorithm. The smaller the maximum value of the delay estimation, the shorter the convergence time of the delay estimation algorithm.
  • the maximum value of the adjusted delay estimation can be set in the delay estimation algorithm module.
  • the maximum value of the time delay estimation is limited to a smaller range, and the real value of the time delay estimated by the time delay estimation algorithm is also within this range, and the final speech signal processing effect will be better.
  • the disclosure can adjust the maximum value of time delay estimation in a targeted manner, and greatly shorten the convergence time of the algorithm.
  • the delay estimation adjustment method is highly pertinent. Users who use each original car can freely initiate adjustments, so that the delay estimation algorithm can adjust the maximum value of the delay estimate in a targeted manner according to the original car used by the user. While greatly shortening the convergence time of the algorithm, it can also obtain a real and effective delay value.
  • FIG. 2 shows a flow 200 of another embodiment of the delay estimation adjustment method according to the present disclosure.
  • the delay estimation adjustment method includes the following steps:
  • Step 201 initialize the maximum value of time delay estimation.
  • step 201 has been introduced in detail in step 101 in the embodiment shown in FIG. 1 , and will not be repeated here.
  • Step 202 Continuously estimate the delay value during the playing of the audition music, obtain at least one delay value of the audition music, and obtain a first delay value based on the at least one delay value of the audition music.
  • the executor of the delay estimation adjustment method may continuously estimate the delay value during the playback of the audition music, obtain at least one delay value of the audition music, and at least one delay value based on the audition music , to get the first delay value.
  • the smart rearview mirror transmits the audition music to the original car machine through screen projection technology, and the original car machine starts to play the audition music.
  • the delay estimation algorithm is used to continuously estimate at least one delay value of the audition music, so as to obtain the first delay value. Since there is a delay jitter in the delay estimation process, the accuracy of an estimated delay value may decrease due to the jitter. Therefore, the first delay value is obtained based on at least one delay value of the audition music, which can improve the accuracy of delay estimation. For example, the average value of at least one time delay value of the audition music is calculated as the first time delay value.
  • Step 203 Continuously estimate the delay value during the playback of the sound test synthesized speech to obtain at least one delay value of the sound test synthesized speech, and at least one delay value based on the sound test synthesized speech to obtain a second delay value.
  • the executor of the delay estimation and adjustment method may continue to estimate the delay value during the playback of the voice test synthesized voice to obtain at least one time delay value of the voice test synthesized voice and at least one time delay value based on the voice test synthesized voice. delay value to obtain the second delay value.
  • the smart rearview mirror transmits the synthesized voice of the test sound to the original car machine through the screen projection technology, and the original car machine starts to play the synthesized voice of the test sound.
  • the time delay estimation algorithm is used to continuously estimate at least one delay value of the synthesized voice of audition, so as to obtain the second delay value. Since there is a delay jitter in the delay estimation process, the accuracy of an estimated delay value may decrease due to the jitter. Therefore, the second time delay value is obtained based on at least one time delay value of the sound-test synthesized speech, which can improve the accuracy of time delay estimation. For example, an average value of at least one time delay value of the sound-test synthesized speech is calculated as the second time delay value.
  • Step 204 Adjust the maximum value of the time delay estimate based on the maximum value of the first time delay value and the second time delay value.
  • the execution subject may adjust the maximum value of the time delay estimate based on the maximum value of the first time delay value and the second time delay value.
  • the maximum value of the delay estimate can be adjusted based on the maximum value of the first delay value and the second delay value.
  • the maximum value of the first delay value and the second delay value is used as the maximum value of the delay estimate.
  • a preset duration is added to the first delay value and the maximum value of the delay value as the maximum value of the delay estimate.
  • the maximum value of the adjusted delay estimation is smaller than the upper limit value of the delay estimation algorithm. The smaller the maximum value of the delay estimation, the shorter the convergence time of the delay estimation algorithm.
  • the maximum value of the adjusted delay estimation can be set in the delay estimation algorithm module. Adjust the maximum value of the delay estimate to the maximum value between the first delay value and the second delay value, regardless of whether the smart rearview mirror transmits music to the original car for playback through screen projection technology, or transmits synthetic voice Playing in the original car machine can completely process the voice interference signal of the voice signal, and the final voice signal processing effect will be better.
  • the delay estimation adjustment method in this embodiment highlights the delay estimation step. Therefore, the solution described in this embodiment continuously estimates the delay value of the audio at least once during the audio playback process. Since there is a delay jitter in the delay estimation process, the accuracy of an estimated delay value may decrease due to the jitter. Therefore, the delay value is obtained based on at least one delay value of the audio, which can improve the accuracy of delay estimation.
  • FIG. 3 shows a flow 300 of another embodiment of the delay estimation adjustment method according to the present disclosure.
  • the delay estimation adjustment method includes the following steps:
  • Step 301 initialize the maximum value of time delay estimation.
  • step 301 has been introduced in detail in step 101 in the embodiment shown in FIG. 1 , and will not be repeated here.
  • step 302 the time delay value is continuously estimated during the playing of the audition music until the same time delay value is obtained for a first preset number of consecutive times as the first time delay value.
  • the executor of the delay estimation adjustment method may continue to estimate the delay value during the sound test music playback until the same delay value is obtained for a first preset number of consecutive times (for example, 5 times), as the second A delay value.
  • the accuracy of an estimated delay value may decrease due to the jitter. If the same time delay value is obtained for the first preset number of consecutive times, it means that there is almost no jitter in the time delay value. Using the same time delay value for the first preset number of consecutive times as the first time delay value can improve the accuracy of time delay estimation.
  • step 303 the time delay value is continuously estimated during the playback of the sound test synthesized voice until the same time delay value is obtained for a second preset number of consecutive times as the second time delay value.
  • the executor of the delay estimation and adjustment method may continue to estimate the delay value during the playback of the sound test synthesized voice until the same delay value is obtained for a second consecutive preset number of times (for example, 5 times), as Second delay value.
  • the accuracy of an estimated delay value may decrease due to the jitter. If the same time delay value is obtained for the second preset number of consecutive times, it means that there is almost no jitter in the time delay value. Using the same time delay value for a second preset number of consecutive times as the second time delay value can improve the accuracy of time delay estimation.
  • first preset number and the second preset number may be the same or different, which are not specifically limited here.
  • step 304 the maximum value of the first delay value and the second delay value is increased by a preset time period as the maximum value of the delay estimate.
  • the execution subject may increase the maximum value of the first delay value and the second delay value by a preset duration as the maximum value of the estimated delay.
  • the maximum value of the adjusted delay estimation can be set in the delay estimation algorithm module. Since there may be delay jitter in the delay estimation process, increasing the maximum value of the first delay value and the second delay value by a preset duration can prevent delay jitter.
  • the delay estimation adjustment method in this embodiment highlights the delay estimation step. Therefore, the solution described in this embodiment continuously estimates the same delay value of the audio for multiple consecutive times during the audio playback process. Since there is a delay jitter in the delay estimation process, the accuracy of an estimated delay value may decrease due to the jitter. Therefore, if the same time delay value is obtained consecutively for many times, it means that there is almost no jitter in the time delay value, thereby improving the accuracy of time delay estimation.
  • FIG. 4 shows a scene diagram in which the delay estimation adjustment method of the embodiment of the present disclosure can be implemented.
  • Step 401 the terminal initiates autonomous adjustment.
  • Step 402 prompting to go online.
  • Step 403 setting the estimated maximum value of time delay as 2500ms.
  • the end user initiates an autonomous adjustment request for the delay estimation, and the voice assistant enters the corresponding mode after being connected, and sets the maximum value of the delay estimation as the upper limit of the algorithm to 2500ms.
  • Step 404 start to play the audition music.
  • Step 405 obtaining the same delay value for 5 consecutive times.
  • the sound test music starts to be played, and the delay estimation algorithm continues to estimate the delay value during the playback process until the same delay value T1 is obtained five times in a row.
  • Step 406 start playing the sound test TTS.
  • Step 407 obtaining the same delay value for 5 consecutive times.
  • Step 408 take the maximum value of the two as the result value.
  • Step 409 add 250ms anti-shake to the result.
  • the comparison shows that the larger value of T1 and T2 is used as T3, and 250ms is added to T3 as T4, which is used as a cache and anti-shake.
  • Step 410 use the new delay result value as the maximum value.
  • T4 is set in the algorithm module as the final maximum value of the delay estimation algorithm of the terminal device.
  • the present disclosure provides an embodiment of a delay estimation adjustment device, which corresponds to the method embodiment shown in FIG. 1 , and the device Specifically, it can be applied to various electronic devices.
  • the apparatus 500 for adjusting delay estimation in this embodiment may include: a setting module 501 , a first estimation module 502 , a second estimation module 503 and an adjustment module 504 .
  • the setting module 501 is configured to initialize the maximum value of the time delay estimation
  • the first estimation module 502 is configured to estimate the first time delay value during the audition music playing process
  • the second estimation module 503 is configured to Estimating the second delay value during the playback of the sound test synthesized voice
  • the adjustment module 504 is configured to adjust the maximum value of the estimated delay based on the first delay value and the second delay value.
  • the first estimating module 502 includes: a first estimating submodule configured to continuously estimate the delay value during the playback of the audition music, and obtain at least one time delay of the audition music. delay value, and at least one time delay value based on the audition music, to obtain the first delay value; and the second estimation module 503 includes: the second estimation submodule, configured to continuously estimate the time delay during the sound audition synthesis speech playback process The delay value is obtained by obtaining at least one delay value of the sound-test synthesized speech, and the second delay value is obtained based on the at least one delay value of the sound-test synthesized speech.
  • the first estimating submodule is further configured to: continue to estimate the delay value during the playback of the audition music until the same delay value is obtained for a first preset number of times in a row , as the first delay value; and the second estimating submodule is further configured to: continue to estimate the delay value during the sound test synthesis voice playback process, until the same delay value is obtained for the second consecutive preset number of times, as the second Second delay value.
  • the adjusting module 504 includes: an adjusting submodule configured to adjust the maximum value of the estimated delay based on the maximum value of the first delay value and the second delay value.
  • the adjustment submodule is further configured to: increase the maximum value of the first delay value and the second delay value by a preset duration as the maximum value of the estimated delay.
  • the setting module 501 is further configured to: initially set the maximum value of the delay estimation as the upper limit value of the delay estimation algorithm.
  • the acquisition, storage and application of the user's personal information involved are in compliance with relevant laws and regulations, and do not violate public order and good customs.
  • the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.
  • FIG. 6 shows a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present disclosure.
  • Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers.
  • Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.
  • the device 600 includes a computing unit 601 that can execute according to a computer program stored in a read-only memory (ROM) 602 or loaded from a storage unit 608 into a random-access memory (RAM) 603. Various appropriate actions and treatments. In the RAM 603, various programs and data necessary for the operation of the device 600 can also be stored.
  • the computing unit 601, ROM 602, and RAM 603 are connected to each other through a bus 604.
  • An input/output (I/O) interface 605 is also connected to the bus 604 .
  • the I/O interface 605 includes: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, an optical disk, etc. ; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, and the like.
  • the communication unit 609 allows the device 600 to exchange information/data with other devices over a computer network such as the Internet and/or various telecommunication networks.
  • Computing unit 601 may be various general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples of computing units 601 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc.
  • the computing unit 601 executes various methods and processes described above, such as a delay estimation adjustment method.
  • the latency estimate adjustment method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608 .
  • part or all of the computer program may be loaded and/or installed on the device 600 via the ROM 602 and/or the communication unit 609.
  • the computer program When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the delay estimation adjustment method described above can be performed.
  • the calculation unit 601 may be configured in any other appropriate way (for example, by means of firmware) to execute the delay estimation adjustment method.
  • Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips Implemented in a system of systems (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof.
  • FPGAs field programmable gate arrays
  • ASICs application specific integrated circuits
  • ASSPs application specific standard products
  • SOC system of systems
  • CPLD load programmable logic device
  • computer hardware firmware, software, and/or combinations thereof.
  • programmable processor can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.
  • Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special purpose computer, or other programmable data processing devices, so that the program codes, when executed by the processor or controller, make the functions/functions specified in the flow diagrams and/or block diagrams Action is implemented.
  • the program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
  • a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device.
  • a machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium.
  • a machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing.
  • machine-readable storage media would include one or more wire-based electrical connections, portable computer discs, hard drives, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read only memory
  • EPROM or flash memory erasable programmable read only memory
  • CD-ROM compact disk read only memory
  • magnetic storage or any suitable combination of the foregoing.
  • the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user. ); and a keyboard and pointing device (eg, a mouse or a trackball) through which a user can provide input to the computer.
  • a display device e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor
  • a keyboard and pointing device eg, a mouse or a trackball
  • Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.
  • the systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system.
  • the components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.
  • a computer system may include clients and servers.
  • Clients and servers are generally remote from each other and typically interact through a communication network.
  • the relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.
  • the server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.
  • steps may be reordered, added or deleted using the various forms of flow shown above.
  • each step described in the present disclosure may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present disclosure can be achieved, no limitation is imposed herein.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Environmental & Geological Engineering (AREA)
  • Mechanical Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Telephone Function (AREA)

Abstract

本公开提供了时延估计调整方法、装置、设备、存储介质以及程序产品，涉及人工智能领域，尤其涉及计算机技术、智能交通领域。具体实现方案为：初始化时延估计的最大值；在试音音乐播放过程中估计第一时延值；在试音合成语音播放过程中估计第二时延值；基于第一时延值和第二时延值，调整时延估计的最大值。本公开能够针对性地调整时延估计的最大值，大大缩短算法收敛时间。

Description

时延估计调整方法、装置、设备以及存储介质
本专利申请要求于2021年6月28日提交的、申请号为202110717530.8、发明名称为“时延估计调整方法、装置、设备以及存储介质”的中国专利申请的优先权，该申请的全文以引用的方式并入本申请中。
技术领域
本公开涉及人工智能领域，尤其涉及计算机技术、智能交通领域。
背景技术
智能后视镜可以通过投屏技术将音频传输到原生车机中进行播放。但因为传输数据量巨大，往往音频播放存在一定的延时。这类延时对语音信号处理产生了较大的负面影响，因为硬件降噪芯片的限制，目前的信号处理技术无法在非常规(>40ms)延时的情况进行。因此，智能后视镜引入了时延估计算法，通过计算时延并对语音信号进行延时。
为了不影响整体体验，时延估计算法通常会把时延估计的最大值限定在较小的范围内(比如1000ms)。但因为每款支持投屏的原生车机性能不一致，导致真实的传输时延波动较大。如果此时将时延估计的最大值统一加大，又会出现收敛时间加长的问题。
发明内容
本公开提供了一种时延估计调整方法、装置、设备、存储介质以及程序产品。
根据本公开的第一方面，提供了一种时延估计调整方法，包括：初始化时延估计的最大值；在试音音乐播放过程中估计第一时延值；在试音合成语音播放过程中估计第二时延值；基于第一时延值和第二时延值，调整时延估计的最大值。
根据本公开的第二方面，提供了一种时延估计调整装置，包括：设 置模块，被配置成初始化时延估计的最大值；第一估计模块，被配置成在试音音乐播放过程中估计第一时延值；第二估计模块，被配置成在试音合成语音播放过程中估计第二时延值；调整模块，被配置成基于第一时延值和第二时延值，调整时延估计的最大值。
根据本公开的第三方面，提供了一种电子设备，包括：至少一个处理器；以及与至少一个处理器通信连接的存储器；其中，存储器存储有可被至少一个处理器执行的指令，指令被至少一个处理器执行，以使至少一个处理器能够执行如第一方面中任一实现方式描述的方法。
根据本公开的第四方面，提供了一种存储有计算机指令的非瞬时计算机可读存储介质，其中，计算机指令用于使计算机执行如第一方面中任一实现方式描述的方法。
根据本公开的第五方面，提供了一种计算机程序产品，包括计算机程序，计算机程序在被处理器执行时实现如第一方面中任一实现方式描述的方法。
本公开能够针对性地调整时延估计的最大值，大大缩短算法收敛时间。
应当理解，本部分所描述的内容并非旨在标识本公开的实施例的关键或重要特征，也不用于限制本公开的范围。本公开的其它特征将通过以下的说明书而变得容易理解。
附图说明
附图用于更好地理解本方案，不构成对本公开的限定。其中：
图1是根据本公开的时延估计调整方法的一个实施例的流程图；
图2是根据本公开的时延估计调整方法的又一个实施例的流程图；
图3是根据本公开的时延估计调整方法的另一个实施例的流程图；
图4是可以实现本公开实施例的时延估计调整方法的场景图；
图5是根据本公开的时延估计调整装置的一个实施例的结构示意图；
图6是用来实现本公开实施例的时延估计调整方法的电子设备的框图。
具体实施方式
以下结合附图对本公开的示范性实施例做出说明，其中包括本公开实施例的各种细节以助于理解，应当将它们认为仅仅是示范性的。因此，本领域普通技术人员应当认识到，可以对这里描述的实施例做出各种改变和修改，而不会背离本公开的范围和精神。同样，为了清楚和简明，以下的描述中省略了对公知功能和结构的描述。
需要说明的是，在不冲突的情况下，本公开中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本公开。
图1示出了根据本公开的时延估计调整方法的一个实施例的流程100。该时延估计调整方法包括以下步骤：
步骤101，初始化时延估计的最大值。
在本实施例中，时延估计调整方法的执行主体可以初始化时延估计的最大值。在后续音频播放过程中，估计出的时延值会受时延估计的最大值的限制，不会超过时延估计的最大值。对于不超过时延估计的最大值的音频，可以对其语音干扰信号彻底处理。对于超过时延估计的最大值的音频，则无法对其语音干扰信号彻底处理。通常，对于任意原生车机(例如车载装置，车载设备，车载软件等)，终端用户可以发起时延估计的自动调整请求，联机后语音助手进入相应模式，初始化时延估计的最大值。为了实现对更多音频的语音干扰信号彻底处理，时延估计的最大值通常会被初始化成一个相对较大的值。在一些实施例中，上述执行主体可以将时延估计的最大值(例如2500ms)设置为时延估计算法的上限值。由于任意音频的时延值都不会超过时延估计算法的上限值，因此将时延估计的最大值初始设置为时延估计算法的上限值，能够实现对任意音频的语音干扰信号彻底处理。
步骤102，在试音音乐播放过程中估计第一时延值。
在本实施例中，上述执行主体可以在试音音乐播放过程中估计第一时延值。
通常，智能后视镜可以通过投屏技术将音乐传输到原生车机中进行 播放。由于音乐数据量巨大，从智能后视镜传输到原生车机中需要一定的时间，导致音乐播放存在一定的延时。因此，这里需要在试音音乐播放过程中估计从智能后视镜传输试音音乐到原生车机播放试音音乐的时延值，即第一时延值。具体地，智能后视镜通过投屏技术将试音音乐传输到原生车机，原生车机开始播放试音音乐。在试音音乐播放过程中，利用时延估计算法估计试音音乐的时延值，以得到第一时延值。例如，利用时延估计算法估计一次试音音乐的时延值，作为第一时延值。其中，试音音乐可以是为了调整时延估计的最大值而播放的任意音乐。试音音乐播放过程就是从智能后视镜传输试音音乐到原生车机播放试音音乐的过程。时延估计算法可以用于估计音频播放的时延值，包括但不限于当前已有的时延估计算法和未来可能出现的时延估计算法。例如，ETDE算法，是一种带约束的时延估计方法，滤波器采用N阶插值运算，直接用瞬时时延估值代替滤波器权系数进行迭代，减小了计算量。
步骤103，在试音合成语音播放过程中估计第二时延值。
在本实施例中，上述执行主体可以在试音合成语音播放过程中估计第二时延值。
通常，智能后视镜可以通过投屏技术将合成语音传输到原生车机中进行播放。由于合成语音数据量巨大，从智能后视镜传输到原生车机中需要一定的时间，导致合成语音播放存在一定的延时。因此，这里需要在试音合成语音播放过程中估计从智能后视镜传输试音合成语音到原生车机播放试音合成语音的时延值，即第二时延值。具体地，智能后视镜通过投屏技术将试音合成语音传输到原生车机，原生车机开始播放试音合成语音。在试音合成语音播放过程中，利用时延估计算法估计试音合成语音的时延值，以得到第二时延值。例如，利用时延估计算法估计一次试音合成语音的时延值，作为第二时延值。其中，合成语音可以是智能后视镜基于采集的信息生成的语音指令。试音合成语音可以是为了调整时延估计的最大值而播放的任意合成语音。试音合成语音播放过程就是从智能后视镜传输试音合成语音到原生车机播放试音合成语音的过程。
步骤104，基于第一时延值和第二时延值，调整时延估计的最大值。
在本实施例中，上述执行主体可以基于第一时延值和第二时延值，调 整时延估计的最大值。
通常，时延估计的最大值可以基于第一时延值和第二时延值中的任意一项进行调整。例如，将第一时延值或第二时延值中的一项作为时延估计的最大值。又例如，将第一时延值或时延值中的一项增加预设时长，作为时延估计的最大值。调整后的时延估计的最大值小于时延估计算法的上限值。时延估计的最大值越小，时延估计算法的收敛时间越短。
这里，调整后的时延估计的最大值可以设置到时延估计算法模块中。将时延估计的最大值限定在较小的范围内，利用时延估计算法估计的时延真实值也在这个范围内，最终的语音信号处理效果也会越好。
本公开能够针对性地调整时延估计的最大值，大大缩短算法收敛时间。时延估计调整方法的针对性强。使用每台原生车机的用户都可以自由地发起调整，使得时延估计算法能够根据用户使用的这台原生车机，进行针对性地调整时延估计的最大值。在大大缩短算法收敛时间的同时，还能够得到真实有效地时延值。
继续参考图2，其示出了根据本公开的时延估计调整方法的又一个实施例的流程200。该时延估计调整方法包括以下步骤：
步骤201，初始化时延估计的最大值。
在本实施例中，步骤201具体操作已在图1所示的实施例中步骤101进行了详细的介绍，在此不再赘述。
步骤202，在试音音乐播放过程中持续估计时延值，得到试音音乐的至少一次时延值，以及基于试音音乐的至少一次时延值，得到第一时延值。
在本实施例中，时延估计调整方法的执行主体可以在试音音乐播放过程中持续估计时延值，得到试音音乐的至少一次时延值，以及基于试音音乐的至少一次时延值，得到第一时延值。
这里，智能后视镜通过投屏技术将试音音乐传输到原生车机，原生车机开始播放试音音乐。在试音音乐播放过程中，利用时延估计算法持续估计试音音乐的至少一次时延值，以得到第一时延值。由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。因此，基于试音音乐的至少一次时延值，得到第一时延值，能够提高时延估计的准确性。例如，计算试音音乐的至少一次时延值的平均值，作为第一时延 值。
步骤203，在试音合成语音播放过程中持续估计时延值，得到试音合成语音的至少一次时延值，以及基于试音合成语音的至少一次时延值，得到第二时延值。
在本实施例中，时延估计调整方法的执行主体可以在试音合成语音播放过程中持续估计时延值，得到试音合成语音的至少一次时延值，以及基于试音合成语音的至少一次时延值，得到第二时延值。
这里，智能后视镜通过投屏技术将试音合成语音传输到原生车机，原生车机开始播放试音合成语音。在试音合成语音播放过程中，利用时延估计算法持续估计试音合成语音的至少一次时延值，以得到第二时延值。由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。因此，基于试音合成语音的至少一次时延值，得到第二时延值，能够提高时延估计的准确性。例如，计算试音合成语音的至少一次时延值的平均值，作为第二时延值。
步骤204，基于第一时延值和第二时延值中的最大值，调整时延估计的最大值。
在本实施例中，上述执行主体可以基于第一时延值和第二时延值中的最大值，调整时延估计的最大值。
通常，时延估计的最大值可以基于第一时延值和第二时延值中的最大值进行调整。例如，将第一时延值和第二时延值中的最大值作为时延估计的最大值。又例如，将第一时延值和时延值中的最大值增加预设时长，作为时延估计的最大值。调整后的时延估计的最大值小于时延估计算法的上限值。时延估计的最大值越小，时延估计算法的收敛时间越短。
这里，调整后的时延估计的最大值可以设置到时延估计算法模块中。将时延估计的最大值调整为第一时延值和第二时延值中的最大值，无论智能后视镜通过投屏技术将音乐传输到原生车机中进行播放，还是将合成语音传输到原生车机中进行播放，均能够对语音信号的语音干扰信号彻底处理，最终的语音信号处理效果也会越好。
从图2中可以看出，与图1对应的实施例相比，本实施例中的时延估计调整方法突出了时延估计步骤。由此，本实施例描述的方案在音频播 放过程中，持续估计音频的至少一次时延值。由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。因此，基于音频的至少一次时延值，得到时延值，能够提高时延估计的准确性。
进一步参考图3，其示出了根据本公开的时延估计调整方法的另一个实施例的流程300。该时延估计调整方法包括以下步骤：
步骤301，初始化时延估计的最大值。
在本实施例中，步骤301具体操作已在图1所示的实施例中步骤101进行了详细的介绍，在此不再赘述。
步骤302，在试音音乐播放过程中持续估计时延值，直至得到连续第一预设数目次相同的时延值，作为第一时延值。
在本实施例中，时延估计调整方法的执行主体可以在试音音乐播放过程中持续估计时延值，直至得到连续第一预设数目次(例如5次)相同的时延值，作为第一时延值。
由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。若得到连续第一预设数目次相同的时延值，说明该时延值几乎不存在抖动。将连续第一预设数目次相同的时延值，作为第一时延值，能够提高时延估计的准确性。
步骤303，在试音合成语音播放过程中持续估计时延值，直至得到连续第二预设数目次相同的时延值，作为第二时延值。
在本实施例中，时延估计调整方法的执行主体可以在试音合成语音播放过程中持续估计时延值，直至得到连续第二预设数目次(例如5次)相同的时延值，作为第二时延值。
由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。若得到连续第二预设数目次相同的时延值，说明该时延值几乎不存在抖动。将连续第二预设数目次相同的时延值，作为第二时延值，能够提高时延估计的准确性。
需要说明的是，第一预设数目与第二预设数目可以相同，也可以不同，这里不进行具体限定。
步骤304，将第一时延值和第二时延值中的最大值增加预设时长，作为时延估计的最大值。
在本实施例中，上述执行主体可以将第一时延值和第二时延值中的最大值增加预设时长，作为时延估计的最大值。
这里，调整后的时延估计的最大值可以设置到时延估计算法模块中。由于时延估计过程中会存在时延抖动，将第一时延值和第二时延值中的最大值增加预设时长，可以防止时延抖动。
从图3中可以看出，与图1对应的实施例相比，本实施例中的时延估计调整方法突出了时延估计步骤。由此，本实施例描述的方案在音频播放过程中，持续估计音频的连续多次相同的时延值。由于时延估计过程中会存在时延抖动，一次估计的时延值可能因存在抖动而准确性下降。因此，若得到连续多次相同的时延值，说明该时延值几乎不存在抖动，从而提高时延估计的准确性。
进一步参考图4，其示出了可以实现本公开实施例的时延估计调整方法的场景图。
步骤401，终端发起自主调整。
步骤402，提示进行联机。
步骤403，将时延估计最大值设置为2500ms。
这里，终端用户发起时延估计的自主调整请求，联机后语音助手进入相应模式，将时延估计的最大值设为算法上限值2500ms。
步骤404，开始播放试音音乐。
步骤405，得出连续5次相同的时延值。
这里，开始播放试音的音乐，时延估计算法在播放过程中持续估算时延值，直到得出连续5次相同的时延值T1。
步骤406，开始播放试音TTS。
步骤407，得到连续5次相同的时延值。
这里，开始播放试音TTS(Text To Speech，从文本到语音)，时延估计算法在播放过程中持续估算时延值，直到得出连续5次相同的时延值T2。
步骤408，取两者中的最大值，作为结果值。
步骤409，将结果增加250ms防抖动。
这里，对比得出T1、T2中较大的值作为T3，在T3基础上增加 250ms为T4，作为缓存和防抖动。
步骤410，使用新时延结果值作为最大值。
这里，将T4作为该终端设备最终的时延估计算法最大值，设置到算法模块中。
进一步参考图5，作为对上述各图所示方法的实现，本公开提供了一种时延估计调整装置的一个实施例，该装置实施例与图1所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。
如图5所示，本实施例的时延估计调整装置500可以包括：设置模块501、第一估计模块502、第二估计模块503和调整模块504。其中，设置模块501，被配置成初始化时延估计的最大值；第一估计模块502，被配置成在试音音乐播放过程中估计第一时延值；第二估计模块503，被配置成在试音合成语音播放过程中估计第二时延值；调整模块504，被配置成基于第一时延值和第二时延值，调整时延估计的最大值。
在本实施例中，时延估计调整装置500中设置模块501、第一估计模块502、第二估计模块503和调整模块504的具体处理及其所带来的技术效果可分别参考图1对应实施例中的步骤101-104的相关说明，在此不再赘述。
在本实施例的一些可选的实现方式中，第一估计模块502包括：第一估计子模块，被配置成在试音音乐播放过程中持续估计时延值，得到试音音乐的至少一次时延值，以及基于试音音乐的至少一次时延值，得到第一时延值；以及第二估计模块503包括：第二估计子模块，被配置成在试音合成语音播放过程中持续估计时延值，得到试音合成语音的至少一次时延值，以及基于试音合成语音的至少一次时延值，得到第二时延值。
在本实施例的一些可选的实现方式中，第一估计子模块进一步被配置成：在试音音乐播放过程中持续估计时延值，直至得到连续第一预设数目次相同的时延值，作为第一时延值；以及第二估计子模块进一步被配置成：在试音合成语音播放过程中持续估计时延值，直至得到连续第二预设数目次相同的时延值，作为第二时延值。
在本实施例的一些可选的实现方式中，调整模块504包括：调整子模块，被配置成基于第一时延值和第二时延值中的最大值，调整时延估计 的最大值。
在本实施例的一些可选的实现方式中，调整子模块进一步被配置成：将第一时延值和第二时延值中的最大值增加预设时长，作为时延估计的最大值。
在本实施例的一些可选的实现方式中，设置模块501进一步被配置成：将时延估计的最大值初始设置为时延估计算法的上限值。
本公开的技术方案中，所涉及的用户个人信息的获取，存储和应用等，均符合相关法律法规的规定，且不违背公序良俗。
根据本公开的实施例，本公开还提供了一种电子设备、一种可读存储介质和一种计算机程序产品。
图6示出了可以用来实施本公开的实施例的示例电子设备600的示意性框图。电子设备旨在表示各种形式的数字计算机，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本公开的实现。
如图6所示，设备600包括计算单元601，其可以根据存储在只读存储器(ROM)602中的计算机程序或者从存储单元608加载到随机访问存储器(RAM)603中的计算机程序，来执行各种适当的动作和处理。在RAM 603中，还可存储设备600操作所需的各种程序和数据。计算单元601、ROM 602以及RAM 603通过总线604彼此相连。输入/输出(I/O)接口605也连接至总线604。
设备600中的多个部件连接至I/O接口605，包括：输入单元606，例如键盘、鼠标等；输出单元607，例如各种类型的显示器、扬声器等；存储单元608，例如磁盘、光盘等；以及通信单元609，例如网卡、调制解调器、无线通信收发机等。通信单元609允许设备600通过诸如因特网的计算机网络和/或各种电信网络与其他设备交换信息/数据。
计算单元601可以是各种具有处理和计算能力的通用和/或专用处 理组件。计算单元601的一些示例包括但不限于中央处理单元(CPU)、图形处理单元(GPU)、各种专用的人工智能(AI)计算芯片、各种运行机器学习模型算法的计算单元、数字信号处理器(DSP)、以及任何适当的处理器、控制器、微控制器等。计算单元601执行上文所描述的各个方法和处理，例如时延估计调整方法。例如，在一些实施例中，时延估计调整方法可被实现为计算机软件程序，其被有形地包含于机器可读介质，例如存储单元608。在一些实施例中，计算机程序的部分或者全部可以经由ROM 602和/或通信单元609而被载入和/或安装到设备600上。当计算机程序加载到RAM 603并由计算单元601执行时，可以执行上文描述的时延估计调整方法的一个或多个步骤。备选地，在其他实施例中，计算单元601可以通过其他任何适当的方式(例如，借助于固件)而被配置为执行时延估计调整方法。
本文中以上描述的系统和技术的各种实施方式可以在数字电子电路系统、集成电路系统、场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、芯片上系统的系统(SOC)、负载可编程逻辑设备(CPLD)、计算机硬件、固件、软件、和/或它们的组合中实现。这些各种实施方式可以包括：实施在一个或者多个计算机程序中，该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释，该可编程处理器可以是专用或者通用可编程处理器，可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令，并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。
用于实施本公开的方法的程序代码可以采用一个或多个编程语言的任何组合来编写。这些程序代码可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器或控制器，使得程序代码当由处理器或控制器执行时使流程图和/或框图中所规定的功能/操作被实施。程序代码可以完全在机器上执行、部分地在机器上执行，作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。
在本公开的上下文中，机器可读介质可以是有形的介质，其可以 包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的程序。机器可读介质可以是机器可读信号介质或机器可读储存介质。机器可读介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备，或者上述内容的任何合适组合。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦除可编程只读存储器(EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。
为了提供与用户的交互，可以在计算机上实施此处描述的系统和技术，该计算机具有：用于向用户显示信息的显示装置(例如，CRT(阴极射线管)或者LCD(液晶显示器)监视器)；以及键盘和指向装置(例如，鼠标或者轨迹球)，用户可以通过该键盘和该指向装置来将输入提供给计算机。其它种类的装置还可以用于提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈(例如，视觉反馈、听觉反馈、或者触觉反馈)；并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如，作为数据服务器)、或者包括中间件部件的计算系统(例如，应用服务器)、或者包括前端部件的计算系统(例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前端部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如，通信网络)来将系统的部件相互连接。通信网络的示例包括：局域网(LAN)、广域网(WAN)和互联网。
计算机系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。服务器可以是云服务器，也可以为分布式系统的服务器，或 者是结合了区块链的服务器。
应该理解，可以使用上面所示的各种形式的流程，重新排序、增加或删除步骤。例如，本发公开中记载的各步骤可以并行地执行也可以顺序地执行也可以不同的次序执行，只要能够实现本公开公开的技术方案所期望的结果，本文在此不进行限制。
上述具体实施方式，并不构成对本公开保护范围的限制。本领域技术人员应该明白的是，根据设计要求和其他因素，可以进行各种修改、组合、子组合和替代。任何在本公开的精神和原则之内所作的修改、等同替换和改进等，均应包含在本公开保护范围之内。

Claims (15)

  1. 一种时延估计调整方法，包括：
    初始化时延估计的最大值；
    在试音音乐播放过程中估计第一时延值；
    在试音合成语音播放过程中估计第二时延值；
    基于所述第一时延值和所述第二时延值，调整所述时延估计的最大值。
  2. 根据权利要求1所述的方法，其中，
    所述在试音音乐播放过程中估计第一时延值，包括：
    在试音音乐播放过程中持续估计时延值，得到所述试音音乐的至少一次时延值，以及基于所述试音音乐的至少一次时延值，得到所述第一时延值；以及
    所述在试音合成语音播放过程中估计第二时延值，包括：
    在试音合成语音播放过程中持续估计时延值，得到所述试音合成语音的至少一次时延值，以及基于所述试音合成语音的至少一次时延值，得到所述第二时延值。
  3. 根据权利要求2所述的方法，其中，
    所述在试音音乐播放过程中持续估计时延值，得到所述试音音乐的至少一次时延值，以及基于所述试音音乐的至少一次时延值，得到所述第一时延值，包括：
    在试音音乐播放过程中持续估计时延值，直至得到连续第一预设数目次相同的时延值，作为所述第一时延值；以及
    所述在试音合成语音播放过程中持续估计时延值，得到所述试音合成语音的至少一次时延值，以及基于所述试音合成语音的至少一次时延值，得到所述第二时延值，包括：
    在试音合成语音播放过程中持续估计时延值，直至得到连续第二预设数目次相同的时延值，作为所述第二时延值。
  4. 根据权利要求1所述的方法，其中，所述基于所述第一时延值和所述第二时延值，调整所述时延估计的最大值，包括：
    基于所述第一时延值和所述第二时延值中的最大值，调整所述时延估计的最大值。
  5. 根据权利要求4所述的方法，其中，所述基于所述第一时延值和所述第二时延值中的最大值，调整所述时延估计的最大值，包括：
    将所述第一时延值和所述第二时延值中的最大值增加预设时长，作为所述时延估计的最大值。
  6. 根据权利要求1-5中任一项所述的方法，其中，所述初始化时延估计的最大值，包括：
    将所述时延估计的最大值初始设置为时延估计算法的上限值。
  7. 一种时延估计调整装置，包括：
    设置模块，被配置成初始化时延估计的最大值；
    第一估计模块，被配置成在试音音乐播放过程中估计第一时延值；
    第二估计模块，被配置成在试音合成语音播放过程中估计第二时延值；
    调整模块，被配置成基于所述第一时延值和所述第二时延值，调整所述时延估计的最大值。
  8. 根据权利要求7所述的装置，其中，
    所述第一估计模块包括：
    第一估计子模块，被配置成在试音音乐播放过程中持续估计时延值，得到所述试音音乐的至少一次时延值，以及基于所述试音音乐的至少一次时延值，得到所述第一时延值；以及
    所述第二估计模块包括：
    第二估计子模块，被配置成在试音合成语音播放过程中持续估计时 延值，得到所述试音合成语音的至少一次时延值，以及基于所述试音合成语音的至少一次时延值，得到所述第二时延值。
  9. 根据权利要求8所述的装置，其中，
    所述第一估计子模块进一步被配置成：
    在试音音乐播放过程中持续估计时延值，直至得到连续第一预设数目次相同的时延值，作为所述第一时延值；以及
    所述第二估计子模块进一步被配置成：
    在试音合成语音播放过程中持续估计时延值，直至得到连续第二预设数目次相同的时延值，作为所述第二时延值。
  10. 根据权利要求7所述的装置，其中，所述调整模块包括：
    调整子模块，被配置成基于所述第一时延值和所述第二时延值中的最大值，调整所述时延估计的最大值。
  11. 根据权利要求10所述的装置，其中，所述调整子模块进一步被配置成：
    将所述第一时延值和所述第二时延值中的最大值增加预设时长，作为所述时延估计的最大值。
  12. 根据权利要求7-11中任一项所述的装置，其中，所述设置模块进一步被配置成：
    将所述时延估计的最大值初始设置为时延估计算法的上限值。
  13. 一种电子设备，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1-6中任一项所述的方法。
  14. 一种存储有计算机指令的非瞬时计算机可读存储介质，其中，所述计算机指令用于使计算机执行根据权利要求1-6中任一项所述的方法。
  15. 一种计算机程序产品，包括计算机程序，所述计算机程序在被处理器执行时实现根据权利要求1-6中任一项所述的方法。
PCT/CN2021/135071 2021-06-28 2021-12-02 时延估计调整方法、装置、设备以及存储介质 WO2023273156A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110717530.8 2021-06-28
CN202110717530.8A CN113382081B (zh) 2021-06-28 2021-06-28 时延估计调整方法、装置、设备以及存储介质

Publications (1)

Publication Number Publication Date
WO2023273156A1 true WO2023273156A1 (zh) 2023-01-05

Family

ID=77579590

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/135071 WO2023273156A1 (zh) 2021-06-28 2021-12-02 时延估计调整方法、装置、设备以及存储介质

Country Status (2)

Country Link
CN (1) CN113382081B (zh)
WO (1) WO2023273156A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113382081B (zh) * 2021-06-28 2023-04-07 阿波罗智联(北京)科技有限公司 时延估计调整方法、装置、设备以及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9246545B1 (en) * 2014-04-11 2016-01-26 Amazon Technologies, Inc. Adaptive estimation of delay in audio systems
US20160134759A1 (en) * 2014-11-06 2016-05-12 Imagination Technologies Limited Pure Delay Estimation
CN107393549A (zh) * 2017-07-21 2017-11-24 北京华捷艾米科技有限公司 时延估计方法及装置
CN109215667A (zh) * 2017-06-29 2019-01-15 华为技术有限公司 时延估计方法及装置
CN113382081A (zh) * 2021-06-28 2021-09-10 阿波罗智联(北京)科技有限公司 时延估计调整方法、装置、设备以及存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102419595B1 (ko) * 2016-01-07 2022-07-11 삼성전자주식회사 재생 지연 조절 방법 및 이를 적용한 전자 장치
US11037583B2 (en) * 2018-08-29 2021-06-15 International Business Machines Corporation Detection of music segment in audio signal
CN110620793B (zh) * 2019-10-31 2022-03-15 苏州浪潮智能科技有限公司 一种提高音频质量的方法、设备及介质
CN110992973A (zh) * 2019-11-29 2020-04-10 维沃移动通信有限公司 一种信号时延的确定方法和电子设备
CN111402868B (zh) * 2020-03-17 2023-10-24 阿波罗智联(北京)科技有限公司 语音识别方法、装置、电子设备及计算机可读存储介质
CN112489670B (zh) * 2020-12-01 2023-08-18 广州华多网络科技有限公司 时延估计方法、装置、终端设备和计算机可读存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9246545B1 (en) * 2014-04-11 2016-01-26 Amazon Technologies, Inc. Adaptive estimation of delay in audio systems
US20160134759A1 (en) * 2014-11-06 2016-05-12 Imagination Technologies Limited Pure Delay Estimation
CN109215667A (zh) * 2017-06-29 2019-01-15 华为技术有限公司 时延估计方法及装置
CN107393549A (zh) * 2017-07-21 2017-11-24 北京华捷艾米科技有限公司 时延估计方法及装置
CN113382081A (zh) * 2021-06-28 2021-09-10 阿波罗智联(北京)科技有限公司 时延估计调整方法、装置、设备以及存储介质

Also Published As

Publication number Publication date
CN113382081A (zh) 2021-09-10
CN113382081B (zh) 2023-04-07

Similar Documents

Publication Publication Date Title
US12057135B2 (en) Speech noise reduction method and apparatus, computing device, and computer-readable storage medium
CN110554357B (zh) 声源定位方法和装置
US20180367674A1 (en) System and method for suppression of non-linear acoustic echoes
WO2020034779A1 (zh) 音频处理方法、存储介质及电子设备
JP7209674B2 (ja) 音声認識方法、音声認識装置、電子機器、コンピュータ可読記憶媒体及びプログラム
WO2018086444A1 (zh) 噪声抑制信噪比估计方法和用户终端
US9196244B2 (en) Methodology for enhanced voice search experience
WO2023273156A1 (zh) 时延估计调整方法、装置、设备以及存储介质
JP2022020060A (ja) ノイズ除去アルゴリズムのデバッグ方法、装置及び電子機器
CN111405336B (zh) 多设备的同步播放方法及系统、电子设备、存储介质
KR102190833B1 (ko) 에코 억제
WO2019080685A1 (zh) 视频图像分割方法及装置、存储介质、电子设备
US20190043497A1 (en) Dynamic adaptation of language understanding systems to acoustic environments
CN111627455A (zh) 一种音频数据降噪方法、装置以及计算机可读存储介质
WO2020107627A1 (zh) 音量调节方法、装置、电子设备及计算机可读存储介质
KR102607120B1 (ko) 소리 데이터 노이즈 캔슬링 방법, 장치, 전자 기기, 컴퓨터 판독 가능 저장 매체 및 컴퓨터 프로그램
JP6471825B1 (ja) 情報処理装置および情報処理方法
CN114299982A (zh) 处理音频信号的方法、装置及电子设备
US10650839B2 (en) Infinite impulse response acoustic echo cancellation in the frequency domain
US20230096150A1 (en) Method and apparatus for determining echo, and storage medium
KR20200078018A (ko) 전자 장치 및 이의 제어 방법
US10897492B1 (en) Delayed VoIP packet delivery
CN114841341A (zh) 模型训练及数据处理方法、装置、设备和存储介质
CN114999440A (zh) 虚拟形象生成方法、装置、设备、存储介质以及程序产品
JP7159884B2 (ja) 情報処理装置および情報処理方法

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: 21948082

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE