WO2017202290A1 - 语音数据传输方法及装置 - Google Patents

语音数据传输方法及装置 Download PDF

Info

Publication number
WO2017202290A1
WO2017202290A1 PCT/CN2017/085457 CN2017085457W WO2017202290A1 WO 2017202290 A1 WO2017202290 A1 WO 2017202290A1 CN 2017085457 W CN2017085457 W CN 2017085457W WO 2017202290 A1 WO2017202290 A1 WO 2017202290A1
Authority
WO
WIPO (PCT)
Prior art keywords
voice
frame
voice data
packet loss
data frame
Prior art date
Application number
PCT/CN2017/085457
Other languages
English (en)
French (fr)
Inventor
梁俊斌
Original Assignee
腾讯科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Priority to EP17802136.6A priority Critical patent/EP3468132A4/en
Publication of WO2017202290A1 publication Critical patent/WO2017202290A1/zh
Priority to US15/984,569 priority patent/US10594449B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • H04L1/188—Time-out mechanisms
    • H04L1/1883—Time-out mechanisms using multiple timers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • H04L1/189—Transmission or retransmission of more than one copy of a message
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M7/00—Arrangements for interconnection between switching centres
    • H04M7/006—Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP), including next generation networks with a packet-switched transport layer
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • H04L1/1887—Scheduling and prioritising arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • H04L1/1874—Buffer management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/16—Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
    • H04L69/164—Adaptation or special uses of UDP protocol

Definitions

  • the present application relates to the field of computer communications, and in particular to a voice data transmission method and a voice data transmission device.
  • Voice over Internet Protocol is a real-time voice communication system based on Ethernet. Network quality and network transmission capability have a decisive influence on the voice quality of Voip. Voip usually uses UDP to ensure real-time data transmission.
  • UDP User Datagram Protocol
  • OSI Open System Interconnection
  • a voice data transmission method including:
  • packet loss retransmission processing is performed only on the voice key frames in the packet loss retransmission processing queue.
  • An example of the present application discloses a voice data transmission apparatus, including:
  • the sending module is configured to put multiple voice data frames into the target sending queue, and send them to the data receiver in turn;
  • the identification insertion module is configured to identify a plurality of voice data frames in the target transmission queue, and put the identified voice key frames into a packet loss retransmission processing queue;
  • a retransmission module configured to perform a packet loss retransmission process only on the voice key frames in the packet loss retransmission processing queue for the multiple voice data frames.
  • An example of the present application discloses a mobile electronic device including a processor, a memory, and an input device and an output device; wherein the processor performs the above-described voice data transmission method by executing a synchronization service management program stored in the memory All steps.
  • An example of the present application discloses a computer storage medium storing a program including all the steps of the above-described voice data transmission method when executed.
  • FIG. 1 is a scenario architecture diagram of a voice data transmission method provided by an example of the present application.
  • FIG. 2 is a schematic flowchart of a voice data transmission method provided by an example of the present application.
  • FIG. 3 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application.
  • FIG. 4 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application.
  • FIG. 5 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application.
  • FIG. 6 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application.
  • FIG. 7 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application.
  • FIG. 8 is a schematic structural diagram of a voice data transmission apparatus provided by an example of the present application.
  • FIG. 9 is a schematic structural diagram of a retransmission module provided by an example of the present application.
  • FIG. 10 is a schematic structural diagram of an identification insertion module provided by an example of the present application.
  • FIG. 11 is a schematic structural diagram of another example of the identification insertion module provided by the present application.
  • FIG. 12 is a schematic structural diagram of another example of the identification insertion module provided by the present application.
  • FIG. 13 is a schematic structural diagram of an analysis unit provided by an example of the present application.
  • FIG. 14 is a structural diagram of an apparatus of an example of the present application.
  • the UDP protocol when used for real-time call data transmission of Voip, packet loss occurs frequently in the case of poor network conditions.
  • the main methods for solving this problem are forward error correction technology and packet loss retransmission technology.
  • the packet loss retransmission is when the data receiver detects that the target data packet has not received the timeout or finds that the target receiving packet is faulty, and sends a request packet to the data sender to request the data sender to retransmit the erroneous data packet.
  • packet loss retransmission mainly includes the following methods:
  • Stopping the packet loss retransmission mode After the data packet is sent, the data sender waits for the status report of the data receiver. If the status report message indicates that the transmission is successful, the subsequent number is sent. According to the message, if the status report message indicates that the transmission failed, the message is retransmitted.
  • Selective retransmission mode When the data sender receives the status message of the data receiver to indicate that the message is in error, the data sender only needs to send the packet with the error.
  • the packet loss retransmission mechanism needs to acknowledge the receiving status of each data packet, that is, it needs to send and receive an acknowledgment message. This operation requires a considerable portion of network bandwidth resources. If the data is received after the first retransmission, the data receiver If the data packet still fails to receive the correct data packet, the data sender needs to continue the retransmission until the success. In the case of weak network capability, each data packet may undergo such a process, which is undoubtedly imposed on the network bandwidth. More pressure is not conducive to data transmission.
  • the present application proposes a voice data transmission method, which is applied to the system architecture shown in FIG. 1.
  • the scene architecture diagram of the voice data transmission method can establish a voice connection between the electronic devices through the network server for voice communication, and then the data sender can collect the collected voice data frames in real time.
  • the data receiver when the data receiver receives the voice data frame and confirms that there is no error, it can return an acknowledgement message indicating that the transmission is successful to the data sender. Otherwise, the data sender can return an acknowledgement message indicating that the transmission failed, and then the data is sent.
  • the party will be able to retransmit the lost voice data frame to the data receiver through the data retransmission mechanism to ensure the quality of the two parties.
  • Step S200 Put multiple voice data frames into the target sending queue, and send them to the data receiver in turn;
  • the data sender establishes a voice communication connection with the data receiver, and the data sender can encode the collected voice information, generate multiple voice data frames, and then put the multiple voice data frames into the target sending queue. And sequentially send the voice data frame to the data receiver.
  • Step S202 Identify a plurality of voice data frames in the target sending queue, and put the identified voice key frames into a packet loss retransmission processing queue;
  • the voice data frame is divided into a voice key frame and a voice non-key frame, and may be a set condition, for example, a voice key frame or a voice non-key frame that meets a preset condition, thereby according to the preset condition.
  • a data frame that does not contain voice content information, or a frame that can be easily recovered by the packet loss compensation method is a voice non-key frame (other than the voice non-key frame)
  • the data frame without voice content information may include non-speech frames or noise frames with minimal energy, and the like.
  • the target sending queue and the packet loss retransmission processing queue in the example of the present application may be different queues, and the target sending queue is configured to normally send a voice data frame to be sent, and the packet loss retransmission processing queue Used to perform retransmission processing when packet loss is performed on voice key frames.
  • the target sending queue and the packet loss retransmission processing queue in the example of the present application may also be the same queue, that is, the voice key frame may be put into the target sending queue, and the voice may be estimated to be resent (ie, retransmitted). The time of the key frame is inserted into the target sending queue, so as to determine that the voice key frame is lost after the packet is lost.
  • Step S204 Perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • the example of the present application performs packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue, and does not perform the packet loss retransmission process on the multiple voice data frames that are sent.
  • the voice non-key frame other than the voice key frame of the frame performs a packet loss retransmission process, that is, whether the voice non-key frame is successfully received by the data receiver, the data sender does not perform the voice non-key frame. Retransmission.
  • the present application performs a retransmission policy by defining a voice non-key frame and determining whether the transmitted voice data frame is a voice non-key frame, and the noise frame of the non-voice frame or the energy technology school is lost.
  • a frame that can be recovered well by the packet compensation method does not perform retransmission, and can save a larger part of the frame retransmission during the call, thereby more effectively utilizing the network bandwidth resource.
  • FIG. 3 a schematic flowchart of another example of the voice data transmission method provided by the present application, as shown in FIG. 3, specifically illustrates how to perform packet loss retransmission processing on a voice key frame in a packet loss retransmission processing queue, including the following steps. :
  • Step S300 Start a first timer to time the current voice key frame in the retransmission transmission window of the packet loss retransmission processing queue.
  • the timing start time of the first timer is a sending time point of the current voice key frame in the target sending queue; the example of the present application is not limited to the time duration of the first timer, and the technician may Experience to set the timing of the first timer.
  • Step S302 detecting whether a status confirmation frame for the current voice key frame returned by the data receiver is received before the first timer expires;
  • the status confirmation frame in the example of the present application indicates that the current voice key frame is successfully received by the data receiver; when the detection result is yes, step S304 is performed, otherwise step S306 is performed.
  • Step S304 deleting the current voice key frame in the packet loss retransmission processing queue
  • Step S306 Retransmit the current voice key frame in the packet loss retransmission processing queue to the data receiver.
  • step S306 it may be returned to step S300 to determine whether the current voice key frame needs to be retransmitted. Further, in order to avoid infinite retransmission of the voice key frame, the example of the present application causes more network bandwidth. With great pressure, you can perform the following steps:
  • the second timer While the first timer is started to be timed, the second timer is started to perform timing, and the timing of the second timer is greater than the timing of the first timer; then, if the second timer is timed before the end of the timer Returning the status confirmation frame to the current voice key frame returned by the data receiver, or deleting the current voice key frame in the packet loss retransmission processing queue when the second timer expires, thereby avoiding Voice keyframes are retransmitted indefinitely, putting more pressure on network bandwidth.
  • the second timer is started to be timed; at this time, the time of the second timer is long, and the technician can according to his own needs or experience. Settings. Receiving a status confirmation frame for the current voice key frame returned by the data receiver before the end of the second timer, or when the second timer expires, in the packet loss retransmission processing queue The current voice key frame is deleted, thereby avoiding infinite retransmission of the voice key frame, which causes more pressure on the network bandwidth.
  • the following describes in detail how to determine whether the current voice data frame is a voice non-key frame in conjunction with FIG. 4 to FIG. 7. It can be understood that the language determined in the multiple voice data frames is The voice data frame other than the non-key frame is the recognized voice key frame:
  • FIG. 4 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application, which may include the following steps:
  • Step S400 Put multiple voice data frames into the target sending queue, and send them to the data receiver in turn;
  • Step S402 Calculating a frame energy of a voice data frame
  • the frame energy Engr of the voice data frame can be calculated by the following formula:
  • Step S404 determining whether the frame energy is less than a first preset threshold
  • the first preset threshold of the example of the present application is related to the value of l. If the value of l is larger, the value of the first preset threshold may be larger. For example, when l is equal to 80, the first preset threshold may be set. 5 * 105; when the determination result is YES, it is judged that the speech data of the speech frame is a non-key frame, perform step S406, otherwise, may be considered as a speech frame of speech data is not non-key frames, step S408.
  • the example of the present application is not limited to the size of the first preset threshold, and the technician may set the first preset threshold according to his own needs or experience.
  • Step S406 Retransmission of the voice data frame is not performed
  • Step S408 Put the identified voice key frame into the packet loss retransmission processing queue
  • Step S410 Perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • step S202 and step S204 in the above example of FIG. 2 may be referred to in step S408 and step S410, and details are not described herein again.
  • FIG. 5 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application, which may include the following steps:
  • Step S500 Put multiple voice data frames into the target sending queue, and send them to the data receiver in turn;
  • Step S502 analyzing whether the voice data frame does not contain voice content information
  • the content of the voice content in the examples in the present application may be understood as the voice data frame does not contain the voice content information at all or the content of the voice content information is lower than a certain threshold; the example of the present application is not limited to analyzing whether the current data frame does not contain The way voice content information is.
  • the analysis result is YES, it is determined that the voice data frame is a voice non-key frame, and step S504 is performed. Otherwise, the voice data frame may be considered as a voice non-key frame, and step S506 is performed.
  • Step S504 Retransmission of the voice data frame is not performed
  • Step S506 Put the identified voice key frame into the packet loss retransmission processing queue
  • Step S508 Perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • step S202 and step S204 in the above example of FIG. 2 may be referred to in step S506 and step S508, and details are not described herein again.
  • FIG. 6 is a schematic flowchart diagram of another example of a voice data transmission method provided by the present application, which may include the following steps:
  • Step S600 Put multiple voice data frames into the target sending queue, and send them to the data receiver in turn;
  • Step S602 Calculate an absolute value of a difference of a coefficient of a first M line pair (1sp) of a previous speech data frame of the speech data frame and the speech data frame;
  • the lsp coefficient of the voice data frame needs to be calculated, for example, by using the voice.
  • Data frame plus Hamming window use Dubin to recursively solve Yule-Walker equation, get llp linear predictive coding (lpc)
  • lpc llp linear predictive coding
  • M can be a positive integer of 5, 6, or 7,
  • M is 6, that is, the example of the present application can compare the absolute values of the first six lsp coefficient differences, and the specific formula can be as follows:
  • the adjacent frame smoothness of the first 6 lsp coefficients of the previous voice data frame of the voice data frame and the voice data frame may be used to determine whether it is a voice non-key frame.
  • Step S604 detecting whether the calculated M absolute values are all smaller than a second preset threshold
  • the second preset threshold of the example of the present application may be 0.005; when the detection result is yes, the voice voice data frame is detected as a voice non-key frame, and step S606 is performed; otherwise, the voice voice data frame may be considered. Step S608 is performed not for the voice non-key frame.
  • the example of the present application is not limited to the size of the second preset threshold, and the technician may set the second preset threshold according to his own needs or experience.
  • Step S606 The voice data frame is not retransmitted
  • Step S608 Put the identified voice key frame into the packet loss retransmission processing queue
  • Step S610 Perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • step S202 and step S204 in the foregoing example of FIG. 2 may be referred to in step S608 and step S610, and details are not described herein again.
  • FIG. 4, FIG. 5 and FIG. 6 can be combined in any way to achieve data weight. Pass, for example:
  • the frame energy of the voice data frame can be first calculated first, and if the energy is less than the first preset threshold, the voice data frame is determined.
  • the voice non-key frame if the energy is not less than the first preset threshold, further analyzing whether the data frame does not contain the voice content information, and when analyzing the voice data frame that does not contain the voice content information, determining the voice data The frame is a voice non-key frame, otherwise it is determined that the voice data frame is not a voice non-key frame.
  • the frame energy of the voice data frame may be first calculated, and if the energy is less than the first preset threshold, the voice data frame is determined.
  • the voice non-key frame if the judgment energy is not less than the first preset threshold, the absolute value of the difference of the first M lsp coefficients of the previous voice data frame of the voice data frame and the voice data frame is further calculated, and the detection is calculated. Whether the M absolute values are smaller than the second preset threshold, and when the detection is less than the second preset threshold, determining that the voice data frame is a voice non-key frame, otherwise determining that the voice data frame is not a voice non-key frame .
  • the data receiver after transmitting the voice data frame to the data receiver, first, it may first analyze whether the voice data frame does not contain voice content information, and when analyzing the voice data frame without voice content information, Determining that the voice data frame is a voice non-key frame, and when analyzing the voice data frame to include voice content information, further calculating a difference of the first M lsp coefficients of the voice data frame and the previous voice data frame opposite to the voice data frame The absolute value is used to detect whether the calculated M absolute values are smaller than the second preset threshold. When the detection is smaller than the second preset threshold, it is determined that the voice data frame is a voice non-key frame, otherwise the voice data frame is determined. Not for voice non-keyframes.
  • FIG. 7 a schematic flowchart of another example of the voice data transmission method provided by the present application, as shown in FIG. 7, may include the following steps:
  • Step S700 Put multiple voice data frames into the target sending queue, and send them sequentially. To the data receiver;
  • Step S702 Calculating a frame energy of the voice data frame
  • step S402 in the above example of FIG. 4, and details are not described herein again.
  • Step S704 determining whether the frame energy is less than a first preset threshold
  • step S718 is performed; otherwise, step S706 is performed.
  • Step S706 performing a 2N point FFFT transform (Fast Fourier Transformation) on the voice data frame to obtain a power spectrum of the voice data frame; and estimating noise energy of each frequency point; the N is a positive integer;
  • MCRA Minima Controlled Recursive Averaging
  • Step S708 Calculate a signal to noise ratio of each frequency point according to the power spectrum and the noise energy
  • Step S710 Calculate a sum of squares of signal to noise ratios of each frequency point according to the signal to noise ratio of each frequency point;
  • Step S712 Compare whether the sum of the squares of the signal-to-noise ratios of the frequency points is smaller than a third preset threshold
  • the third preset threshold of the example of the present application is related to the value of N.
  • the first preset threshold may be set to 2; when the comparison result is yes, step S718 is performed; otherwise, step S718 is performed; otherwise, Step S714 is performed;
  • the example of the present application is not limited to the size of the third preset threshold, the technical person The third preset threshold can be set according to his own needs or experience.
  • Step S714 Calculate an absolute value of a difference of the first M lsp coefficients of the previous voice data frame of the voice data frame and the voice data frame;
  • step S602 in the above example of FIG. 6, and details are not described herein again.
  • Step S716 detecting whether the calculated M absolute values are all smaller than a second preset threshold
  • step S718 is performed, otherwise step S720 is performed.
  • Step S718 The voice data frame is not retransmitted
  • Step S720 Put the identified voice key frame into the packet loss retransmission processing queue
  • Step S722 Perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • step S202 and step S204 in the above example of FIG. 2 may be referred to in step S720 and step S722, and details are not described herein again.
  • any combination of the examples of FIG. 4, FIG. 5 and FIG. 6 is not limited to the above four types, and there are other ways of judging the combination, for example, the order of execution in the combination changes, etc., not all listed here.
  • Combinations that can be implemented by those skilled in the art according to the description of the examples of the present application are all within the scope of the present application.
  • the data receiver can ignore the voice non-key frame according to the actual situation or recover by using a packet loss compensation method, for example, by using the voice.
  • the previous frame with a large correlation of data frames, or a frame with a small difference between the pitch period and the previous frame is restored.
  • multiple voice data frames are placed in a target sending queue, and are sequentially sent to a data receiver; multiple voice data frames in the target sending queue are identified, and the recognized voice key frames are identified.
  • the network bandwidth pressure is increased, which is not conducive to the technical problem of data transmission.
  • the embodiment of the present invention performs a retransmission policy by defining a voice non-key frame and determining whether the transmitted voice data frame is a voice non-key frame. Or the noise frame of the energy technology school or the frame that can be recovered well by the packet loss compensation method does not perform retransmission, which can save a larger part of the frame retransmission during the call, thereby more effectively utilizing the network bandwidth resource.
  • the present application further provides a voice data transmission device, which is a schematic structural diagram of a voice data transmission device provided by the example of the present application, as shown in FIG.
  • the method may include: placing the sending module 800, the identifying putting module 802, and the retransmission module 804, where
  • the sending module 800 is configured to put a plurality of voice data frames into the target sending queue, and send the data frames to the data receivers in sequence;
  • the identification insertion module 802 is configured to identify a plurality of voice data frames in the target transmission queue, and put the identified voice key frames into a packet loss retransmission processing queue;
  • the retransmission module 804 is configured to perform packet loss retransmission processing only on the voice key frames in the packet loss retransmission processing queue for the plurality of voice data frames.
  • the retransmission module 804 may include: a first initiating unit 8040, a first detecting unit 8042, a first deleting unit 8044, and a retransmission subunit. 8046, of which
  • the first starting unit 8040 is configured to start a first timer for timing the current voice key frame in the retransmission sending window of the packet loss retransmission processing queue; the timing start time of the first timer is the current a transmission time point of the voice key frame in the target transmission queue;
  • the first detecting unit 8042 is configured to detect whether a status confirmation frame for the current voice key frame returned by the data receiver is received before the first timer expires;
  • the first deleting unit 8044 is configured to: when the detection result is yes, delete the current voice key frame in the packet loss retransmission processing queue;
  • the retransmission sub-unit 8046 is configured to retransmit the current voice key frame in the packet loss retransmission processing queue to the data receiver when the detection result is no.
  • the retransmission module 804 may further include a second activation unit and a second deletion unit.
  • the second activation unit is configured to start the second time when the first startup unit 8040 starts the first timer to perform timing.
  • the timer is timed, the timing of the second timer is greater than the timing of the first timer, and the second deleting unit is configured to receive the data reception before the second timer expires.
  • the status confirmation frame returned by the party for the current voice key frame, or the current voice key frame is deleted in the packet loss retransmission processing queue when the second timer expires.
  • the retransmission module 804 may further include a third activation unit and a third deletion unit.
  • the third activation unit is configured to retransmit the voice key frame in the packet loss retransmission processing queue to the retransmission subunit 8046.
  • the second timer is started to perform timing; and the third deleting unit is configured to receive, after the end of the second timer, the state returned by the data receiver for the current voice key frame.
  • the current voice key frame is deleted in the packet loss retransmission processing queue.
  • the retransmission module 804 can include the second activation unit, the second deletion unit, and the third activation unit and the third deletion unit.
  • the identification insertion module 802 may include: a first calculation unit 8020 and an energy judgment Breaking unit 8022, wherein
  • the first calculating unit 8020 is configured to calculate a frame energy of the voice data frame
  • the energy determining unit 8022 is configured to determine whether the frame energy is less than a first preset threshold
  • the voice data frame is a voice non-key frame.
  • the identification putting module 802 may further include an analyzing unit 8024, configured to analyze whether the voice data frame does not contain voice content information; when the analysis result is yes, identify the voice data frame as a voice non-key frame.
  • an analyzing unit 8024 configured to analyze whether the voice data frame does not contain voice content information; when the analysis result is yes, identify the voice data frame as a voice non-key frame.
  • the identification insertion module 802 may include: a second calculation unit 8026 and a second detection unit 8028, wherein
  • the second calculating unit 8026 is configured to calculate an absolute value of a difference of the first M lsp coefficients of the previous voice data frame of the voice data frame and the voice data frame;
  • the second detecting unit 8028 is configured to detect whether the calculated M absolute values are all smaller than a second preset threshold
  • the voice data frame is identified as a voice non-key frame.
  • the identification insertion module 802 includes a first calculation unit 8020, an energy determination unit 8022, and an analysis module 8024.
  • a second calculating unit 8026 and a second detecting unit 8028 are included, wherein
  • the trigger analysis module 8024 analyzes whether the voice data frame does not contain the voice content information, and when it is determined that the voice data frame contains the voice content information, the trigger is triggered.
  • the second calculating unit 8026 calculates an absolute value of the difference of the first M lsp coefficients of the previous speech data frame of the speech data frame and the speech data frame.
  • the identification into the module 802, the first computing unit 8020 and the energy is not limited to the above, and any combination of groups may be performed as described in the method examples.
  • the analysis unit 8024 may include: a transformation estimation unit 80240, a signal to noise ratio calculation unit 80242, a square sum calculation unit 80244, and a comparison unit 80246, wherein ,
  • the transform estimating unit 80240 is configured to perform a 2N point fft transform on the voice data frame to obtain a power spectrum of the voice data frame; and estimate noise energy of each frequency point; the N is a positive integer;
  • the signal to noise ratio calculation unit 80242 is configured to calculate a signal to noise ratio of each frequency point according to the power spectrum and the noise energy;
  • the sum of squares calculation unit 80244 is configured to calculate a sum of squares of signal to noise ratios of each frequency point according to the signal to noise ratio of each frequency point;
  • the comparing unit 80246 is configured to compare whether the sum of squares of signal to noise ratios of the frequency points is less than a third preset threshold
  • the voice data frame is identified as a voice non-key frame.
  • the embodiment of the present invention implements multiple voice data frames in a target sending queue, and sequentially sends them to a data receiver; identifies multiple voice data frames in the target sending queue, and identifies them.
  • the outbound voice key frames are placed in the packet loss retransmission processing queue; for multiple voice data frames, the packet loss retransmission process is performed only on the voice key frames in the packet loss retransmission processing queue, and the speech non-key frames are not lost.
  • the packet retransmission process realizes that it is not necessary to confirm the reception status of each voice data frame to see if retransmission is needed, which effectively solves the problem that the prior art has increased network bandwidth pressure due to data retransmission under the condition of weak network capability. The technical problem is not conducive to data transmission.
  • the embodiment of the present invention performs a retransmission policy by defining a voice non-key frame and determining whether the transmitted voice data frame is a voice non-key frame, and a noise frame for a non-voice frame or an energy technical school or Frames that can be recovered well by packet loss compensation, do not perform retransmission, can Save a larger portion of the frame retransmission during the call, thus making more efficient use of network bandwidth resources.
  • each module and unit may exist physically separately, or may be integrated by two or more devices or modules.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the voice data transmission device described above can be executed in various computing devices that can perform user information processing based on the Internet and loaded in the memory of the computing device.
  • Figure 14 is a diagram showing the composition of a computing device in which a voice data transmission device is located.
  • the computing device includes one or more processors (CPUs) 1402, communication modules 1404, memory 1406, user interface 1410, and a communication bus 1408 for interconnecting these components.
  • processors CPUs
  • communication modules 1404, memory 1406, user interface 1410, and a communication bus 1408 for interconnecting these components.
  • the processor 1402 can receive and transmit data through the communication module 1404 to effect network communication and/or local communication.
  • User interface 1410 includes one or more output devices 1412 that include one or more speakers and/or one or more visual displays.
  • User interface 1410 also includes one or more input devices 1414 including, for example, a keyboard, a mouse, a voice command input unit or loudspeaker, a touch screen display, a touch sensitive tablet, a gesture capture camera or other input button or control, and the like.
  • the memory 1406 can be a high speed random access memory such as DRAM, SRAM, DDR RAM, or other random access solid state storage device; or a nonvolatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, Or other non-volatile solid-state storage devices.
  • a high speed random access memory such as DRAM, SRAM, DDR RAM, or other random access solid state storage device
  • nonvolatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, Or other non-volatile solid-state storage devices.
  • the memory 1406 stores a set of instructions executable by the processor 1402, including:
  • Operating system 1416 including for processing various basic system services and for performing hardware phases Procedure for closing tasks;
  • the application 1418 includes various programs for implementing voice data transmission, and the program can implement the processing flow in each of the above examples, and may include, for example, the voice data transmission device 80 shown in FIG.
  • voice data transmission 80 can include modules 802, 804, 806, units 8020, 8022, 8024, 8026, 8028 and units 8040, 8042, 8044, 8046 shown in Figures 8-13, each of which can be The storage machine executes the instructions.
  • the processor 1402 can implement the functions of the above modules and units by executing the machine executable instructions in the modules and units in the memory 1406.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).
  • the present application therefore also discloses a non-volatile storage medium in which is stored a data processing program for performing any of the above-described methods of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Computer Security & Cryptography (AREA)

Abstract

本申请公开了语音数据传输方法，包括：将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。本申请还公开了语音数据传输装置。

Description

语音数据传输方法及装置
本申请要求于2016年05月25日提交中国专利局、申请号为201610356121.9、发明名称为“一种语音数据传输方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机通信领域，尤其一种语音数据传输方法以及语音数据传输装置。
背景
IP承载语音(Voice over Internet Protocol，Voip)是基于以太网的实时语音通话系统，网络质量、网络传输能力对Voip双方通话质量有着决定性的影响，Voip为了确保实时通话数据传输，通常采用的是UDP(User Datagram Protocol，用户数据包协议)协议，所述UDP协议是OSI(Open System Interconnection，开放式系统互联)参考模型中一种无连接的传输层协议，提供面向事务的简单不可靠信息传送服务，IETF RFC 768是UDP的正式规范。
技术内容
为了解决上述技术问题，本申请的实例公开了一种语音数据传输方法，包括：
将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
本申请的实例公开了一种语音数据传输装置，包括：
放入发送模块，用于将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
识别放入模块，用于对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
重传模块，用于针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
本申请的实例公开了一种移动电子设备，包括处理器、存储器以及输入装置和输出装置；其中，所述处理器通过执行所述存储器中存储的同步服务管理程序，执行上述语音数据传输方法的全部步骤。
本申请的实例公开了一种计算机存储介质，所述计算机存储介质存储有程序，所述程序执行时包括上述语音数据传输方法的全部步骤。
附图简要说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实例提供的语音数据传输方法的场景架构图；
图2是本申请实例提供的语音数据传输方法的流程示意图；
图3是本申请提供的语音数据传输方法的另一实例的流程示意图；
图4是本申请提供的语音数据传输方法的另一实例的流程示意图；
图5是本申请提供的语音数据传输方法的另一实例的流程示意图；
图6是本申请提供的语音数据传输方法的另一实例的流程示意图；
图7是本申请提供的语音数据传输方法的另一实例的流程示意图；
图8是本申请实例提供的语音数据传输装置的结构示意图；
图9是本申请实例提供的重传模块的结构示意图；
图10是本申请实例提供的识别放入模块的结构示意图；
图11是本申请提供的识别放入模块的另一实例的结构示意图；
图12是本申请提供的识别放入模块的另一实例的结构示意图；
图13是本申请实例提供的分析单元的结构示意图；以及
图14是本申请一实例的设备结构图。
实施方式
下面将结合本申请实例中的附图，对本申请实例中的技术方案进行清楚、完整地描述，显然，所描述的实例仅仅是本发明一部分实例，而不是全部的实例。基于本申请中的实例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实例，都属于本申请保护的范围。
在一些实例中，采用UDP协议进行Voip的实时通话数据传输时，在网络状况不佳情况下丢包是经常发生的，解决该问题的主要方法有前向纠错技术和丢包重传技术。
其中，丢包重传是当数据接收方检测到目标数据包超时仍未接收到或者发现目标接收包出错，则向数据发送方发出请求包，以请求数据发送方重传出错的数据报文的一种技术手段。在一些实例中，丢包重传主要包括以下几种方式：
a、停等式丢包重传方式：数据报文发送完成后，数据发送方等待数据接收方的状态报告，如果状态报告报文表示发送成功才发送后续的数 据报文，如果状态报告报文表示发送失败则重传该报文。
b、回退N帧丢包重传方式：当数据发送方接收到数据接收方的状态报文指示报文出错后，发送方将重传过去N个报文。
c、选择性重传方式：当数据发送方接收到数据接收方的状态报文指示报文出错后，数据发送方只需发送发生错误的报文。
上述丢包重传机制需要每个数据报文都需要做接收状态确认，即需要发送和接收确认报文，这种操作需要占用相当一部分网络带宽资源，如果第一次重传后，数据接收方仍然没有成功接收到正确的数据报文，则数据发送方需要继续进行重传直到成功为止，在网络能力较弱情况下，每个数据报文都可能经历这样的过程，无疑是对网络带宽施加更大的压力，不利于进行数据传输。
为了解决上述技术问题，本申请提出一种语音数据传输方法，该方法应用于图1所示的系统构架中。
如图1示出的本申请实例提供的语音数据传输方法的场景架构图，电子设备之间可以通过网络服务器建立语音连接，进行语音通信，那么数据发送方可以实时地将采集到的语音数据帧发送给数据接收方，数据接收方接收到语音数据帧并确认没有出错时，可以向数据发送方返回指示发送成功的确认消息，否则可以向数据发送方返回指示发送失败的确认消息，那么数据发送方将可以通过数据重传机制，对丢失的语音数据帧重传给数据接收方，以保证双方通话质量。
下面结合图2示出的本申请实例提供的语音数据传输方法的流程示意图，详细说明本发明实施例提供的语音数据传输方法的技术方案，包括如下步骤：
步骤S200：将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
具体地，数据发送方与数据接收方建立了语音通信连接，该数据发送方可以将采集到的语音信息进行编码，生成多个语音数据帧，然后将该多个语音数据帧放入目标发送队列，并依次将语音数据帧发送给数据接收方。
步骤S202：对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
具体地，本申请实例中对语音数据帧分为语音关键帧和语音非关键帧，可以通过设定条件，例如符合预设条件的为语音关键帧或语音非关键帧，从而根据该预设条件来对该多个语音数据帧进行识别，例如可以设定不含语音内容信息的数据帧，或者容易通过丢包补偿手段能很好恢复的帧为语音非关键帧(语音非关键帧以外的即为语音关键帧)，该不含语音内容信息的数据帧可以包括非语音帧或能量极小的噪声帧，等等。当从该多个语音数据帧中识别出语音关键帧后，将该语音关键帧放入预设的丢包重传处理队列。
需要说明的是，本申请实例中的目标发送队列与丢包重传处理队列可以为不同的两个队列，该目标发送队列用于正常发送待发送的语音数据帧，该丢包重传处理队列用于对语音关键帧执行丢包时的重传处理。本申请实例中的目标发送队列与丢包重传处理队列也可以为同一个队列，也就是说，可以将语音关键帧放入目标发送队列中，具体可以估算再次发送(即重传)该语音关键帧的时间来插入到目标发送队列中，以便于确定该语音关键帧发生丢包后，进行重传。
步骤S204：针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，本申请实例对发送的多个语音数据帧，仅仅对处于该丢包重传处理队列中的语音关键帧执行丢包重传处理，不对该多个语音数据 帧的语音关键帧以外的语音非关键帧执行丢包重传处理，也就是说，无论该语音非关键帧是否被该数据接收方成功接收到，该数据发送方都不对该语音非关键帧进行重传。
通过实施本申请实例，实现了无需对每个语音数据帧都做接收状态确认看是否需要进行重传，有效解决了在网络能力较弱情况下因数据重传导致网络带宽压力增大，不利于解决网络丢包的技术问题，本申请实例通过定义语音非关键帧，并判断发送的语音数据帧是否为语音非关键帧来执行重传策略，对非语音帧或能量技校的噪声帧或通过丢包补偿手段能很好恢复的帧，不执行重传，可以节省通话过程中较大一部分的帧重传，从而更加有效地利用了网络带宽资源。
具体地，如图3示出的本申请提供的语音数据传输方法的另一实例的流程示意图，具体说明如何对丢包重传处理队列中的语音关键帧执行丢包重传处理，包括如下步骤：
步骤S300：针对处于所述丢包重传处理队列的重传发送窗口的当前语音关键帧，启动第一定时器进行计时；
具体地，该第一定时器的计时起算时间为该当前语音关键帧在该目标发送队列中的发送时间点；本申请实例不限于该第一定时器的计时时长，技术人员可以根据自身需求或经验来设置该第一定时器的计时时长。
步骤S302：检测是否在所述第一定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧；
具体地，本申请实例中的状态确认帧表明该当前语音关键帧被数据接收方成功接收到；当检测结果为是时，则执行步骤S304，否则执行步骤S306。
步骤S304：在所述丢包重传处理队列中将所述当前语音关键帧删除；
步骤S306：将所述丢包重传处理队列中的所述当前语音关键帧重传给所述数据接收方。
具体地，步骤S306之后，可以重新返回执行步骤S300，再次判断该当前语音关键帧是否需要重传；进一步地，本申请实例为了避免对语音关键帧进行无限次的重传，给网络带宽造成更大压力，可以执行如下步骤：
在启动第一定时器进行计时的同时，启动第二定时器进行计时，该第二定时器的计时时长大于该第一定时器的计时时长；那么，若在该第二定时器计时结束前接收到数据接收方返回的针对所述当前语音关键帧的状态确认帧，或者在该第二定时器计时结束时，在该丢包重传处理队列中将该当前语音关键帧删除，从而避免了对语音关键帧进行无限次的重传，给网络带宽造成更大压力。
或者，
在将该丢包重传处理队列中的语音关键帧重传给该数据接收方之后，启动第二定时器进行计时；此时第二定时器的计时时长，技术人员可以根据自身需求或经验来设置。在该第二定时器计时结束前接收到该数据接收方返回的针对所述当前语音关键帧的状态确认帧时，或者在该第二定时器计时结束时，在该丢包重传处理队列中将所述当前语音关键帧删除，从而避免了对语音关键帧进行无限次的重传，给网络带宽造成更大压力。
具体地，下面结合图4至图7详细举例说明如何判断当前语音数据帧是否为语音非关键帧，可理解的是，该多个语音数据帧中判断出的语 音非关键帧以外的语音数据帧，即为识别出的语音关键帧：
如图4示出的本申请提供的语音数据传输方法的另一实例的流程示意图，可以包括如下步骤：
步骤S400：将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
步骤S402：计算语音数据帧的帧能量；
具体地，可以通过以下公式，计算出该语音数据帧的帧能量Engr：
Figure PCTCN2017085457-appb-000001
其中，l为帧长
步骤S404：判断该帧能量是否小于第一预设阈值；
具体地，本申请实例的第一预设阈值与l的取值有关，l越大，该第一预设阈值的取值可以越大，例如l等于80时，该第一预设阈值可以设为5*105；当判断结果为是时，则判断出该语音数据帧为语音非关键帧，执行步骤S406，否则，可以认为该语音数据帧不为语音非关键帧，执行步骤S408。
需要说明的是，本申请实例不限于该第一预设阈值的大小，技术人员可以根据自身需求或经验来设置该第一预设阈值。
步骤S406：不对所述语音数据帧进行重传；
步骤S408：将识别出的语音关键帧放入丢包重传处理队列；
步骤S410：针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，步骤S408和步骤S410可以参考上述图2实例中步骤S202和步骤S204的描述，这里不再赘述。
如图5示出的本申请提供的语音数据传输方法的另一实例的流程示意图，可以包括如下步骤：
步骤S500：将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
步骤S502：分析语音数据帧是否不含有语音内容信息；
具体地，本申请各个实例中的不含有语音内容信息可以理解为语音数据帧完全不含有语音内容信息或者含有的语音内容信息量低于某阈值；本申请实例不限于分析当前数据帧是否不含有语音内容信息的方式。当分析结果为是时，则判断出该语音数据帧为语音非关键帧，执行步骤S504，否则，可以认为该语音数据帧不为语音非关键帧，执行步骤S506。
步骤S504：不对所述语音数据帧进行重传；
步骤S506：将识别出的语音关键帧放入丢包重传处理队列；
步骤S508：针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，步骤S506和步骤S508可以参考上述图2实例中步骤S202和步骤S204的描述，这里不再赘述。
如图6示出的本申请提供的语音数据传输方法的另一实例的流程示意图，可以包括如下步骤：
步骤S600：将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
步骤S602：计算语音数据帧与该语音数据帧相对的上一个语音数据帧的前M个线谱对(Linear Spectrum Pair，lsp)系数差的绝对值；
具体地，在计算该语音数据帧与该语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值之前，还需要计算出语音数据帧的lsp系数，例如可以通过对该语音数据帧加汉明窗，使用dubin递推求解Yule-Walker方程，得到lpc线性预测编码(linear predictive coding，lpc) 系数，然后根据得到的p阶lpc系数，求解下列求解p+1阶对称多项式P(z)和反对称多项式Q(z)的共轭复根，即可得到lsp系数A：
P(z)＝A(z)+z-(p+1)×A(z-1)
Q(z)＝A(z)+z-(p+1)×A(z-1)
得到lsp系数后，考虑到前三组线谱对对语音影响最重要，常规语音编解码中可以计算得到10个lsp系数，在本申请的实例中M可以为5、6、7等正整数，例如，M为6，也就是说，本申请实例可以比较前6个lsp系数差的绝对值，具体公式可以如下：
lspdif(i)＝abs(lspcur(i)-lsopre(i))；i＝0～5
语音数据帧与该语音数据帧相对的上一个语音数据帧的前6个lsp系数的相邻帧平稳度，可以用以判断是否为语音非关键帧。
步骤S604：检测计算得到的M个绝对值是否都小于第二预设阈值；
具体地，本申请实例的第二预设阈值可以为0.005；当检测结果为是时，则检测出该语音语音数据帧为语音非关键帧，执行步骤S606，否则，可以认为该语音语音数据帧不为语音非关键帧，执行步骤S608。
需要说明的是，本申请实例不限于该第二预设阈值的大小，技术人员可以根据自身需求或经验来设置该第二预设阈值。
步骤S606：不对该语音数据帧进行重传；
步骤S608：将识别出的语音关键帧放入丢包重传处理队列；
步骤S610：针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，步骤S608和步骤S610可以参考上述图2实例中步骤S202和步骤S204的描述，这里不再赘述。
需要说明的是，图4、图5和图6实例可以任意组合来实现数据重 传，例如：
1、可以图4和图5结合：将语音数据帧发送给数据接收方后，首先可以先计算该语音数据帧的帧能量，若判断能量小于第一预设阈值，则判断出该语音数据帧为语音非关键帧，若判断能量不小于第一预设阈值，再进一步分析所述数据帧是否不含有语音内容信息，当分析该语音数据帧不含有语音内容信息时，则判断出该语音数据帧为语音非关键帧，否则判断出该语音数据帧不为语音非关键帧。
2、或者图4和图6结合：将语音数据帧发送给数据接收方后，首先可以先计算该语音数据帧的帧能量，若判断能量小于第一预设阈值，则判断出该语音数据帧为语音非关键帧，若判断能量不小于第一预设阈值，再进一步计算该语音数据帧与该语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值，检测计算得到的M个绝对值是否都小于第二预设阈值，当检测都小于第二预设阈值，则判断出该语音数据帧为语音非关键帧，否则判断出该语音数据帧不为语音非关键帧。
3、又或者图5和图6结合：将语音数据帧发送给数据接收方后，首先可以先分析该语音数据帧是否不含有语音内容信息，当分析语音数据帧不含有语音内容信息时，则判断出该语音数据帧为语音非关键帧，当分析该语音数据帧含有语音内容信息时，再进一步计算语音数据帧与该语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值，检测计算得到的M个绝对值是否都小于第二预设阈值，当检测都小于第二预设阈值，则判断出该语音数据帧为语音非关键帧，否则判断出该语音数据帧不为语音非关键帧。
4、又或者如图4、图5和图6的结合，如图7示出的本申请提供的语音数据传输方法的另一实例的流程示意图，可以包括如下步骤：
步骤S700：将多个语音数据帧放入目标发送队列，并将其依次发送 给数据接收方；
步骤S702：计算语音数据帧的帧能量；
具体地，参考上述图4实例中的步骤S402，这里不再赘述。
步骤S704：判断该帧能量是否小于第一预设阈值；
具体地，参考上述图4实例中的步骤S404，当判断结果为是时，则执行步骤S718；否则执行步骤S706。
步骤S706：对该语音数据帧进行2N点fft变换(Fast Fourier Transformation，快速傅立叶变换)，得到该语音数据帧的功率谱；并估算各频点噪声能量；所述N为正整数；
具体地，步骤S706至步骤S712为分析该语音数据帧是否不含有语音内容信息的其中一个具体实施方式；对该语音数据帧进行2N点fft变换，得到该语音数据帧的功率谱level(j)，j＝1～N；然后可以基于最小统计量控制递归平均算法(Minima Controlled Recursive Averaging，MCRA)算法估计各频点噪声能量noise(j)，j＝1～N。
步骤S708：根据所述功率谱和所述噪声能量计算各频点信噪比；
具体地，各频点信噪比snr(j)＝level(j)/noise(j)，j＝1～N。
步骤S710：根据所述各频点信噪比计算各频点信噪比平方之和；
具体地，各频点信噪比平方之和
Figure PCTCN2017085457-appb-000002
步骤S712：比较所述各频点信噪比平方之和是否小于第三预设阈值；
具体地，本申请实例的的第三预设阈值与N的取值有关，例如N等于64时，该第一预设阈值可以设为2；当比较结果为是时，则执行步骤S718，否则执行步骤S714；
需要说明的是，本申请实例不限于该第三预设阈值的大小，技术人 员可以根据自身需求或经验来设置该第三预设阈值。
步骤S714：计算该语音数据帧与该语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值；
具体地，参考上述图6实例中的步骤S602，这里不再赘述。
步骤S716：检测计算得到的M个绝对值是否都小于第二预设阈值；
具体地，参考上述图6实例中的步骤S604，当检测结果为是时，则执行步骤S718，否则执行步骤S720。
步骤S718：不对该语音数据帧进行重传；
步骤S720：将识别出的语音关键帧放入丢包重传处理队列；
步骤S722：针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，步骤S720和步骤S722可以参考上述图2实例中步骤S202和步骤S204的描述，这里不再赘述。
需要说明的是，本申请实例中图4、图5和图6实例可以任意组合不限于以上4种，还有其它判断组合的方式，例如组合中执行的先后顺序发生变化等，这里不全部罗列，本领域技术人员根据本申请实例的描述即可实现的组合方式都属于本申请的保护范围之内。
还需要说明的是，数据发送方向数据接收方发送的语音非关键帧发生丢包时，数据接收方可以根据实际情况忽略该语音非关键帧或通过丢包补偿手段来恢复，例如通过在该语音数据帧相关性较大的前一帧，或者基音周期与前一帧相差很小的帧来进行恢复。
实施本发明实施例，将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；对目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；针对多个语音数据帧，仅对在丢包重传处理队列中的语音关键帧执行丢包重传处理，而 不对语音非关键帧执行丢包重传处理，实现了无需对每个语音数据帧都做接收状态确认看是否需要进行重传，有效解决了现有技术在网络能力较弱情况下因数据重传导致网络带宽压力增大，不利于进行数据传输的技术问题，本发明实施例通过定义语音非关键帧，并判断发送的语音数据帧是否为语音非关键帧来执行重传策略，对非语音帧或能量技校的噪声帧或通过丢包补偿手段能很好恢复的帧，不执行重传，可以节省通话过程中较大一部分的帧重传，从而更加有效地利用了网络带宽资源。
为了便于更好地实施本申请实例的上述方案，本申请还对应提供了一种语音数据传输装置，如图8示出的本申请实例提供的语音数据传输装置的结构示意图，语音数据传输装置80可以包括：放入发送模块800、识别放入模块802和重传模块804，其中，
放入发送模块800用于将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
识别放入模块802用于对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
重传模块804用于针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
具体地，如图9示出的本申请实例提供的重传模块的结构示意图，重传模块804可以包括：第一启动单元8040、第一检测单元8042、第一删除单元8044和重传子单元8046，其中，
第一启动单元8040用于针对处于所述丢包重传处理队列的重传发送窗口的当前语音关键帧，启动第一定时器进行计时；所述第一定时器的计时起算时间为所述当前语音关键帧在所述目标发送队列中的发送时间点；
第一检测单元8042用于检测是否在所述第一定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧；
第一删除单元8044用于当检测结果为是时，在所述丢包重传处理队列中将所述当前语音关键帧删除；
重传子单元8046用于当检测结果为否时，则将所述丢包重传处理队列中的所述当前语音关键帧重传给所述数据接收方。
进一步地，重传模块804还可以包括第二启动单元和第二删除单元，具体地，该第二启动单元，用于在第一启动单元8040启动第一定时器进行计时的同时，启动第二定时器进行计时，所述第二定时器的计时时长大于所述第一定时器的计时时长；该第二删除单元，用于若在所述第二定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧，或者在所述第二定时器计时结束时，在所述丢包重传处理队列中将所述当前语音关键帧删除。
或者，
重传模块804还可以包括第三启动单元和第三删除单元，具体地，第三启动单元，用于在重传子单元8046将所述丢包重传处理队列中的语音关键帧重传给所述数据接收方之后，启动第二定时器进行计时；第三删除单元，用于在所述第二定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧时，或者在所述第二定时器计时结束时，在所述丢包重传处理队列中将所述当前语音关键帧删除。
需要说明的是，重传模块804可以同时包括第二启动单元、第二删除单元以及第三启动单元和第三删除单元。
再进一步地，如图10示出的本申请实例提供的识别放入模块的结构示意图，识别放入模块802可以包括：第一计算单元8020和能量判 断单元8022，其中，
第一计算单元8020用于计算语音数据帧的帧能量；
能量判断单元8022用于判断所述帧能量是否小于第一预设阈值；
当判断结果为是时，则识别出所述语音数据帧为语音非关键帧。
再进一步地，识别放入模块802还可以包括分析单元8024，用于分析语音数据帧是否不含有语音内容信息；当分析结果为是时，则识别出所述语音数据帧为语音非关键帧。
再进一步地，如图11示出的本申请提供的识别放入模块的另一实例的结构示意图，识别放入模块802可以包括：第二计算单元8026和第二检测单元8028，其中，
第二计算单元8026用于计算语音数据帧与所述语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值；
第二检测单元8028用于检测计算得到的M个绝对值是否都小于第二预设阈值；
当检测结果为是时，则识别出所述语音数据帧为语音非关键帧。
再进一步地，如图12示出的本申请提供的识别放入模块的另一实例的结构示意图，识别放入模块802包括第一计算单元8020、能量判断单元8022和分析模块8024外，还可以包括第二计算单元8026和第二检测单元8028，其中，
能量判断单元8024判断出所述帧能量不小于第一预设阈值时，触发分析模块8024分析语音数据帧是否不含有语音内容信息，当分析出所述语音数据帧含有语音内容信息确定时，触发第二计算单元8026计算所述语音数据帧与所述语音数据帧相对的上一语音数据帧的前M个lsp系数差的绝对值。
需要说明的是，识别放入模块802中，第一计算单元8020和能量 判断单元8022，与分析模块8024，以及第二计算单元8026和第二检测单元8028的组合不限于以上所述，还可以按照方法实例的描述进行任意组的组合。
再进一步地，如图13示出的本申请实例提供的分析单元的结构示意图，分析单元8024可以包括：变换估算单元80240、信噪比计算单元80242、平方和计算单元80244和比较单元80246，其中，
变换估算单元80240用于对所述语音数据帧进行2N点fft变换，得到所述语音数据帧的功率谱；并估算各频点噪声能量；所述N为正整数；
信噪比计算单元80242用于根据所述功率谱和所述噪声能量计算各频点信噪比；
平方和计算单元80244用于根据所述各频点信噪比计算各频点信噪比平方之和；
比较单元80246用于比较所述各频点信噪比平方之和是否小于第三预设阈值；
当比较结果为是时，则识别出所述语音数据帧为语音非关键帧。
综上所述，实施本发明实施例，将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；对目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；针对多个语音数据帧，仅对在丢包重传处理队列中的语音关键帧执行丢包重传处理，而不对语音非关键帧执行丢包重传处理，实现了无需对每个语音数据帧都做接收状态确认看是否需要进行重传，有效解决了现有技术在网络能力较弱情况下因数据重传导致网络带宽压力增大，不利于进行数据传输的技术问题，本发明实施例通过定义语音非关键帧，并判断发送的语音数据帧是否为语音非关键帧来执行重传策略，对非语音帧或能量技校的噪声帧或通过丢包补偿手段能很好恢复的帧，不执行重传，可以 节省通话过程中较大一部分的帧重传，从而更加有效地利用了网络带宽资源。
另外，在本申请各个实例中的方法和装置以及其中的各模块及单元可以集成在一个处理单元中，也可以是各个模块及单元单独物理存在，也可以两个或两个以上装置或模块集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
在一实例中，上述的语音数据传输装置可运行在各种可基于互联网而进行用户信息处理的计算设备中，并加载在该计算设备的存储器中。
图14示出了语音数据传输装置所在的计算设备的组成结构图。如图14所示，该计算设备包括一个或者多个处理器(CPU)1402、通信模块1404、存储器1406、用户接口1410，以及用于互联这些组件的通信总线1408。
处理器1402可通过通信模块1404接收和发送数据以实现网络通信和/或本地通信。
用户接口1410包括一个或多个输出设备1412，其包括一个或多个扬声器和/或一个或多个可视化显示器。用户接口1410也包括一个或多个输入设备1414，其包括诸如，键盘，鼠标，声音命令输入单元或扩音器，触屏显示器，触敏输入板，姿势捕获摄像机或其他输入按钮或控件等。
存储器1406可以是高速随机存取存储器，诸如DRAM、SRAM、DDR RAM、或其他随机存取固态存储设备；或者非易失性存储器，诸如一个或多个磁盘存储设备、光盘存储设备、闪存设备，或其他非易失性固态存储设备。
存储器1406存储处理器1402可执行的指令集，包括：
操作系统1416，包括用于处理各种基本系统服务和用于执行硬件相 关任务的程序；
应用1418，包括用于实现语音数据传输的各种程序，这种程序能够实现上述各实例中的处理流程，比如可以包括图8所示的语音数据传输装置80。
在一些实例中，语音数据传输80可包括图8-13所示的各模块802、804、806、单元8020、8022、8024、8026、8028及单元8040、8042、8044、8046，上述各模块可以存储有机器可执行指令。处理器1402通过执行存储器1406中各模块及单元中的机器可执行指令，进而能够实现上述各模块及单元的功能。
本领域普通技术人员可以理解实现上述实例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实例的流程。其中，所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)或随机存储记忆体(Random Access Memory，RAM)等。
因此本申请还公开了一种非易失性存储介质，其中存储有数据处理程序，该数据处理程序用于执行本申请上述方法的任何一种实例。
以上所揭露的仅为本申请的实例而已，当然不能以此来限定本申请之权利范围，因此依本申请权利要求所作的等同变化，仍属本申请所涵盖的范围。

Claims (17)

  1. 一种语音数据传输方法，包括：
    将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
    对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
    针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
  2. 如权利要求1所述的方法，其中，所述对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理，包括：
    针对处于所述丢包重传处理队列的重传发送窗口的当前语音关键帧，启动第一定时器进行计时；所述第一定时器的计时起算时间为所述当前语音关键帧在所述目标发送队列中的发送时间点；
    检测是否在所述第一定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧；
    当检测结果为是时，则在所述丢包重传处理队列中将所述当前语音关键帧删除；
    当检测结果为否时，则将所述丢包重传处理队列中的所述当前语音关键帧重传给所述数据接收方。
  3. 如权利要求2所述的方法，其中，所述启动第一定时器进行计时的同时，还包括：
    启动第二定时器进行计时，所述第二定时器的计时时长大于所述第一定时器的计时时长；
    若在所述第二定时器计时结束前接收到所述数据接收方返回的针 对所述当前语音关键帧的状态确认帧，或者在所述第二定时器计时结束时，在所述丢包重传处理队列中将所述当前语音关键帧删除。
  4. 如权利要求2所述的方法，其中，所述将所述丢包重传处理队列中的语音关键帧重传给所述数据接收方之后，还包括：
    启动第二定时器进行计时；
    在所述第二定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧时，或者在所述第二定时器计时结束时，在所述丢包重传处理队列中将所述当前语音关键帧删除。
  5. 如权利要求1所述的方法，其中，所述对所述目标发送队列中的多个语音数据帧进行识别，包括：
    计算语音数据帧的帧能量；
    判断所述帧能量是否小于第一预设阈值；
    当判断结果为是时，则识别出所述语音数据帧为语音非关键帧。
  6. 如权利要求1所述的方法，其中，所述对所述目标发送队列中的多个语音数据帧进行识别，包括：
    分析语音数据帧是否不含有语音内容信息；
    当分析结果为是时，则识别出所述语音数据帧为语音非关键帧。
  7. 如权利要求1所述的方法，其中，所述对所述目标发送队列中的多个语音数据帧进行识别，包括：
    计算语音数据帧与所述语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值；
    检测计算得到的M个绝对值是否都小于第二预设阈值；
    当检测结果为是时，则识别出所述语音数据帧为语音非关键帧。
  8. 如权利要求5所述的方法，其中，当判断出所述帧能量不小于第一预设阈值时，还包括：
    分析所述语音数据帧是否不含有语音内容信息；
    当分析结果为是时，则识别出所述语音数据帧为语音非关键帧。
  9. 如权利要求5所述的方法，其中，当判断出所述帧能量不小于第一预设阈值时，还包括：
    计算所述语音数据帧与所述语音数据帧相对的上一语音数据帧的前M个lsp系数差的绝对值；
    检测计算得到的M个绝对值是否都小于第二预设阈值；
    当检测结果为是时，则识别出所述语音数据帧为语音非关键帧。
  10. 如权利要求6或8所述的方法，其中，当分析出所述语音数据帧含有语音内容信息时，还包括：
    计算所述语音数据帧与所述语音数据帧相对的上一语音数据帧的前M个lsp系数差的绝对值；
    检测计算得到的M个绝对值是否都小于第二预设阈值；
    当检测结果为是时，则识别出所述语音数据帧为语音非关键帧。
  11. 如权利要求6或8所述的方法，其中，所述分析所述语音数据帧是否不含有语音内容信息，包括：
    对所述语音数据帧进行2N点fft变换，得到所述语音数据帧的功率谱；并估算各频点噪声能量；所述N为正整数；
    根据所述功率谱和所述噪声能量计算各频点信噪比；
    根据所述各频点信噪比计算各频点信噪比平方之和；
    比较所述各频点信噪比平方之和是否小于第三预设阈值；
    当比较结果为是时，则识别出所述当前语音数据帧为语音非关键帧。
  12. 一种语音数据传输装置，其中，该装置包括处理器和存储器， 所述存储器中存储有计算机可读指令，可以使所述处理器执行以下操作：
    将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
    对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
    针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
  13. 如权利要求12所述的装置，其中，所述计算机可读指令可以使所述处理器执行以下操作：
    针对处于所述丢包重传处理队列的重传发送窗口的当前语音关键帧，启动第一定时器进行计时；所述第一定时器的计时起算时间为所述当前语音关键帧在所述目标发送队列中的发送时间点；
    检测是否在所述第一定时器计时结束前接收到所述数据接收方返回的针对所述当前语音关键帧的状态确认帧；
    当检测结果为是时，在所述丢包重传处理队列中将所述当前语音关键帧删除；
    当检测结果为否时，则将所述丢包重传处理队列中的所述当前语音关键帧重传给所述数据接收方。
  14. 如权利要求12所述的装置，其中，所述计算机可读指令可以使所述处理器执行以下操作：
    计算语音数据帧的帧能量；
    判断所述帧能量是否小于第一预设阈值；
    当判断结果为是时，则识别出所述语音数据帧为语音非关键帧。
  15. 如权利要求12所述的装置，其中，所述计算机可读指令可以 使所述处理器执行以下操作：
    分析语音数据帧是否不含有语音内容信息；
    当分析结果为是时，则识别出所述语音数据帧为语音非关键帧。
  16. 如权利要求12所述的装置，其中，所述计算机可读指令可以使所述处理器执行以下操作：
    计算语音数据帧与所述语音数据帧相对的上一个语音数据帧的前M个lsp系数差的绝对值；
    检测计算得到的M个绝对值是否都小于第二预设阈值；
    当检测结果为是时，则识别出所述语音数据帧为语音非关键帧。
  17. 一种非易失性计算机可读存储介质，存储有计算机可读指令，可以使至少一个处理器执行以下操作：
    将多个语音数据帧放入目标发送队列，并将其依次发送给数据接收方；
    对所述目标发送队列中的多个语音数据帧进行识别，并将识别出的语音关键帧放入丢包重传处理队列；
    针对所述多个语音数据帧，仅对在所述丢包重传处理队列中的语音关键帧执行丢包重传处理。
PCT/CN2017/085457 2016-05-25 2017-05-23 语音数据传输方法及装置 WO2017202290A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP17802136.6A EP3468132A4 (en) 2016-05-25 2017-05-23 METHOD AND DEVICE FOR TRANSMITTING LANGUAGE DATA
US15/984,569 US10594449B2 (en) 2016-05-25 2018-05-21 Voice data transmission method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610356121.9A CN106067847B (zh) 2016-05-25 2016-05-25 一种语音数据传输方法及装置
CN201610356121.9 2016-05-25

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/984,569 Continuation US10594449B2 (en) 2016-05-25 2018-05-21 Voice data transmission method and device

Publications (1)

Publication Number Publication Date
WO2017202290A1 true WO2017202290A1 (zh) 2017-11-30

Family

ID=57420304

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/085457 WO2017202290A1 (zh) 2016-05-25 2017-05-23 语音数据传输方法及装置

Country Status (4)

Country Link
US (1) US10594449B2 (zh)
EP (1) EP3468132A4 (zh)
CN (1) CN106067847B (zh)
WO (1) WO2017202290A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112350816A (zh) * 2020-11-12 2021-02-09 海能达通信股份有限公司 一种数据传输方法、存储介质和通信设备

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106067847B (zh) 2016-05-25 2019-10-22 腾讯科技（深圳）有限公司 一种语音数据传输方法及装置
CN108616925B (zh) * 2016-12-13 2023-03-21 中兴通讯股份有限公司 一种数据流的处理方法及装置
CN108234420B (zh) * 2016-12-21 2021-03-19 北京酷我科技有限公司 一种解决流媒体丢失视频头的系统及方法
CN107277423B (zh) * 2017-07-28 2020-10-09 北京小米移动软件有限公司 丢包重传的方法及装置
CN111371534B (zh) * 2020-06-01 2020-09-18 腾讯科技（深圳）有限公司 一种数据重传方法、装置、电子设备和存储介质
CN113936669A (zh) * 2020-06-28 2022-01-14 腾讯科技（深圳）有限公司 数据传输方法、系统、装置、计算机可读存储介质及设备
CN113936670A (zh) * 2020-06-28 2022-01-14 腾讯科技（深圳）有限公司 丢包重发方法、系统、装置、计算机可读存储介质及设备
CN113037440B (zh) * 2021-05-25 2021-08-06 腾讯科技（深圳）有限公司 数据重传处理方法、装置、计算机设备和存储介质
US11922026B2 (en) 2022-02-16 2024-03-05 T-Mobile Usa, Inc. Preventing data loss in a filesystem by creating duplicates of data in parallel, such as charging data in a wireless telecommunications network
CN116112126B (zh) * 2023-04-12 2023-06-09 深圳市云天智能通讯有限公司 基于通话场景自适应降噪方法、装置及通信终端

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101098173A (zh) * 2006-06-30 2008-01-02 华为技术有限公司 动态调整无线网络中传输的业务服务质量的方法
CN103259803A (zh) * 2013-06-04 2013-08-21 腾讯科技（深圳）有限公司 一种音频回放方法、装置和系统
CN103490972A (zh) * 2013-09-27 2014-01-01 迈普通信技术股份有限公司 多链路隧道报文传输方法及系统
CN104506287A (zh) * 2014-12-29 2015-04-08 重庆邮电大学 一种td-lte应急通信下的实时语音通信方法
WO2015101402A1 (en) * 2013-12-31 2015-07-09 Telecom Italia S.P.A. Congestion management in a multicast communication network
CN106067847A (zh) * 2016-05-25 2016-11-02 腾讯科技（深圳）有限公司 一种语音数据传输方法及装置

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2765715B1 (fr) * 1997-07-04 1999-09-17 Sextant Avionique Procede de recherche d'un modele de bruit dans des signaux sonores bruites
DE60020672T2 (de) * 2000-03-02 2005-11-10 Matsushita Electric Industrial Co., Ltd., Kadoma Verfahren und Vorrichtung zur Wiederholung der Videodatenrahmen mit Prioritätsstufen
CA2396213A1 (en) * 2002-07-31 2004-01-31 Anthony Gerkis Automatic retransmit request protocol for channels with time-varying capacity
CN1271593C (zh) * 2004-12-24 2006-08-23 北京中星微电子有限公司 一种语音信号检测方法
US8260609B2 (en) * 2006-07-31 2012-09-04 Qualcomm Incorporated Systems, methods, and apparatus for wideband encoding and decoding of inactive frames
JP4842075B2 (ja) * 2006-09-28 2011-12-21 京セラ株式会社 音声伝送装置
CN102137438A (zh) * 2010-01-25 2011-07-27 华为技术有限公司 一种ip网络资源的分配方法及装置
EP2529502A1 (en) * 2010-01-28 2012-12-05 Thomson Licensing A method and apparatus for retransmission decision making
US9143279B2 (en) * 2013-01-18 2015-09-22 Qualcomm Incorporated Methods and devices for facilitating data retransmissions in wireless communication systems
US9741350B2 (en) * 2013-02-08 2017-08-22 Qualcomm Incorporated Systems and methods of performing gain control
GB2520536B (en) * 2013-11-25 2017-05-03 Canon Kk Method and device for data communication in a communication network
CN103632677B (zh) * 2013-11-27 2016-09-28 腾讯科技(成都)有限公司 带噪语音信号处理方法、装置及服务器
JPWO2016068316A1 (ja) * 2014-10-31 2017-08-10 日本電気株式会社 無線基地局、パケット送信装置、無線端末、制御方法、及びプログラム
EP3292712B1 (en) * 2015-05-06 2019-11-06 Telefonaktiebolaget LM Ericsson (publ) Methods and devices for handling data packet transmissions in a multi-path multi-hop adapted wireless communication network
GB2539277A (en) * 2015-06-12 2016-12-14 Canon Kk Backoff based selection method of channels for data transmission

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101098173A (zh) * 2006-06-30 2008-01-02 华为技术有限公司 动态调整无线网络中传输的业务服务质量的方法
CN103259803A (zh) * 2013-06-04 2013-08-21 腾讯科技（深圳）有限公司 一种音频回放方法、装置和系统
CN103490972A (zh) * 2013-09-27 2014-01-01 迈普通信技术股份有限公司 多链路隧道报文传输方法及系统
WO2015101402A1 (en) * 2013-12-31 2015-07-09 Telecom Italia S.P.A. Congestion management in a multicast communication network
CN104506287A (zh) * 2014-12-29 2015-04-08 重庆邮电大学 一种td-lte应急通信下的实时语音通信方法
CN106067847A (zh) * 2016-05-25 2016-11-02 腾讯科技（深圳）有限公司 一种语音数据传输方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3468132A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112350816A (zh) * 2020-11-12 2021-02-09 海能达通信股份有限公司 一种数据传输方法、存储介质和通信设备

Also Published As

Publication number Publication date
EP3468132A1 (en) 2019-04-10
EP3468132A4 (en) 2019-10-16
US20180278372A1 (en) 2018-09-27
CN106067847A (zh) 2016-11-02
CN106067847B (zh) 2019-10-22
US10594449B2 (en) 2020-03-17

Similar Documents

Publication Publication Date Title
WO2017202290A1 (zh) 语音数据传输方法及装置
Paxson et al. Computing TCP's retransmission timer
US7496036B2 (en) Method and apparatus for determining client-perceived server response time
Paxson et al. RFC 6298: Computing TCP's retransmission timer
CN113037440B (zh) 数据重传处理方法、装置、计算机设备和存储介质
US8478890B2 (en) System and method for reliable virtual bi-directional data stream communications with single socket point-to-multipoint capability
JP2016213811A (ja) インタラクティブなリアルタイムメディアの転送プロトコル
US20150244650A1 (en) Efficient packet processing at video receiver in multimedia communications over packet networks
WO2023093879A1 (zh) 数据传输方法、装置、设备和介质
WO2020119347A1 (zh) 一种消息传输方法、装置、设备及介质
CN112436924B (zh) 一种数据传输方法及电子设备
CN103763073A (zh) 一种丢包重传的方法及终端
WO2019128964A1 (zh) 一种实现丢包检测的方法、装置和系统
WO2019100475A1 (zh) 一种数据传输方法、装置、设备及存储介质
CN114285830B (zh) 语音信号处理方法、装置、电子设备及可读存储介质
US10129163B2 (en) Methods and apparatus for preventing head of line blocking for RTP over TCP
CN111654505B (zh) Udp数据包可靠传输方法、系统、计算机设备及存储介质
EP1235414A1 (en) Data transmitting apparatus and data receiving apparatus
CN108540273B (zh) 一种数据包重传的方法和装置
WO2023078222A1 (zh) 数据传输方法、装置、设备和介质
CN117676695A (zh) Tcp传输方法、装置和系统
Zhang et al. How edge computing and initial congestion window affect latency of Web-based services: Early experiences with Baidu?
US7643503B2 (en) System and method for dynamically determining retransmit buffer time
US20150012792A1 (en) Method and apparatus for providing a transmission control protocol minimum retransmission timer
JP2016019198A (ja) 通信装置、通信装置の制御方法、プログラム

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 17802136

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2017802136

Country of ref document: EP

Effective date: 20190102