JP2007158875A - Sound transmission apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sound transmission apparatus to reduce the fluctuations of the sound reproduction of a plurality of spots, when the sound is reproduced simultaneously in the plural spots. <P>SOLUTION: A sound transmission terminal 3 transmits a sound multicast data to an IP network 2 and a plurality of sound-receiving terminals 4 receive the sound multicast data through the IP network 2. The sound transmission terminal 3 is provided with a sound packet transmitting portion 12 to transmit a sound packet of the sound multicast data and a learning packet transmitting portion 13 for transmitting a learning packet, including a header constitution similar to the sound packet, before the transmission of the sound packet is started. Each sound-receiving terminal 4 is provided with a learning packet canceling portion 15 for canceling the learning packet, received before the start of the reception of the sound packet. A network path is decided by transmitting the learning packet, and then the transmission of the sound packet is started. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an audio transmission apparatus that transmits audio digital data by multicast via an IP network.

  Conventionally, as an audio transmission apparatus in a premises such as a building or a large-scale facility, an analog audio transmission apparatus has been generally used, but with the development of network technology, a digital audio transmission apparatus has been proposed. . In the case of the digital format, typically, a voice transmitting terminal is connected to a number of voice receiving terminals via an Ethernet (registered trademark) IP network. Audio is digitized on the transmission side, and audio stream data is transmitted to a plurality of audio receiving terminals via the IP network, and audio is reproduced at each audio receiving terminal and output from a speaker. By using network technology, there are advantages such as the ability to freely specify the audio output location.

  In an audio transmission apparatus using an IP network, multicast transmission is advantageously used to send an audio stream to a plurality of audio receiving terminals. For example, an audio stream is transmitted by multicast to many audio receiving terminals in one area, and the same audio is output simultaneously in that area. In such audio transmission, output timings at a plurality of audio receiving terminals are required to be aligned.

As a conventional technique for aligning the output timing of stream data transmitted via an IP network, output timing information is incorporated into the stream. The output timing is controlled according to the output timing information, which makes it difficult for output timing deviations between receiving terminals to occur. Such a technique is found in MPEG2 and the like. Further, this type of technology is disclosed in Patent Document 1. Patent Document 1 discloses a technique for attaching time data to video and audio stream data and synchronizing a plurality of streams based on the time data.
JP 2002-369163 A (FIG. 1 etc.)

  However, in the conventional audio transmission apparatus, if audio is to be reproduced simultaneously by a plurality of audio receiving terminals, an apparatus that incorporates output timing information such as a time code as described above into a stream is required. There is a problem that it gets bigger.

  In order to avoid such a problem, it is not necessary to actively control the output timing without providing a configuration for incorporating a time code or the like into a stream. However, if the control of the output timing is simply stopped, the output timing is arbitrarily determined by the reception timing of the audio data. Since the reception timing varies between audio receiving terminals, the output timing also varies. As a result, fluctuation (jitter) occurs in audio reproduction between speakers.

  The reception timing shift between the voice receiving terminals is caused by the following mechanism. When transmission of an audio stream is started, data is transmitted in a state where a network path for stream transmission has not yet been determined. In the process of determining the network route, a difference occurs between terminals in the time until the network route is determined (hereinafter referred to as network route determination time) due to the learning time of various switches such as hubs constituting the network. The reception start timing of audio data is shifted between terminals. Due to such a shift, there is a possibility that the output timing between adjacent speakers shifts, fluctuations occur, and the sound level shifts.

  In order to prevent the fluctuation caused by the network route determination time as described above, it is also conceivable to use a protocol such as IGMP (Internet Group Management Protocol) which is a technology for determining a network route. However, in this case, there is a problem that a device corresponding to a protocol such as IGMP is required and the cost of the system increases. In particular, in an audio transmission apparatus such as a large-scale facility, the number of speakers is enormous and the number of audio receiving terminals is enormous. It is practically difficult to give high functions to such a huge number of terminals.

  As described above, conventionally, if a configuration in which output timing information is incorporated into a stream or a protocol-compatible device such as IGMP is not provided, reception timing between audio receiving terminals is shifted, and fluctuation in audio reproduction occurs. There was a problem.

  The present invention has been made to solve the conventional problems, and an object of the present invention is to provide an audio transmission device capable of reducing fluctuations in audio reproduction at a plurality of locations when audio is reproduced at a plurality of locations simultaneously. There is.

  The voice transmission device of the present invention includes a voice transmission terminal that sends voice multicast data to an IP network, and a plurality of voice reception terminals that receive the voice multicast data transmitted from the voice transmission terminal through the IP network, The voice transmitting terminal multicasts a learning packet including a voice packet transmitting means for transmitting a voice packet of the voice multicast data and a header structure similar to the voice packet before the voice packet transmission is started. Learning packet transmitting means, and each of the plurality of voice receiving terminals includes learning packet discarding means for discarding the learning packet received before the start of reception of the voice packet.

  With this configuration, since the learning packet is transmitted before the voice packet, the network path of the IP network is determined by the transmission of the learning packet, and then the voice packet passes through the determined network path. Since the voice packet passes through the already determined network path, fluctuation of the voice reception start time between voice receiving terminals is reduced. Therefore, even when audio reproduction is performed at a plurality of locations, fluctuations in audio reproduction at a plurality of locations can be reduced.

  The voice transmission apparatus of the present invention is configured to be able to specify a time interval between transmission of the learning packet by the learning packet transmission unit and transmission start of the voice packet by the voice transmission unit.

  With this configuration, considering the fact that the network route determination time varies depending on the configuration and equipment of each network, start the voice packet so that the transmission of the voice packet starts after the network route is determined by the transmission of the learning packet The timing can be set appropriately and the fluctuation of voice generation can be suitably reduced.

  In the present invention, learning packets are transmitted before voice packets to guide the network route, so that fluctuations in the voice reception start time between voice receiving terminals are reduced, and voice reproduction is performed at a plurality of locations. Even in such a case, it is possible to provide an audio transmission device having an effect that fluctuations in audio reproduction at a plurality of locations can be reduced.

  Hereinafter, an audio transmission apparatus according to an embodiment of the present invention will be described with reference to the drawings.

  An audio transmission / reception system including an audio transmission apparatus according to the first embodiment of the present invention is shown in FIGS. FIG. 1 is a block diagram, and FIG. 2 shows the overall configuration.

  First, the overall configuration of FIG. 2 will be described. The voice transmission device 1 includes a voice transmission terminal 3 and a plurality of voice reception terminals 4, which are connected by an IP network 2, and the IP network 2, voice transmission, and the like. The terminal 3 and the voice receiving terminal 4 constitute a voice transmission / reception system.

  The sound transmitting terminal 3 is provided with a sound source 5. The sound source 5 is, for example, a microphone or a sound reproduction device, and inputs analog or digital sound. The audio reproduction device is, for example, a CD reproduction device. The sound source 5 may be connected to the voice transmission terminal 3 or may be integrated with the voice transmission terminal 3. When voice is input from the sound source 5, the voice / voice transmission terminal 3 generates an IP packet of voice data and sends it to the IP network 2. The audio data is suitably transmitted using multicast for a plurality of audio receiving terminals 4.

  Here, in this embodiment, voice data transmitted by multicast is called voice multicast data, and an IP packet (IP multicast packet) of voice multicast data is called a voice packet.

  The voice packet is received by each voice receiving terminal 4, and analog voice is reproduced by each voice receiving terminal 4. Then, the sound is output from the speaker 6 provided in each sound receiving terminal 4. The speaker 6 may be separate from the voice receiving terminal 4 or may be integrated.

  The voice transmission device 1 in FIG. 2 is provided in a premises such as a building or a large-scale commercial facility, for example. In FIG. 2, some of the voice receiving terminals 4 are shown, but in reality, more voice receiving terminals 4 are preferably provided. For example, in a large facility, the number of voice receiving terminals 4 is enormous, which may range from thousands to tens of thousands. Moreover, although one voice transmission terminal 3 is shown in FIG. 2, a plurality of voice transmission terminals 3 may be provided.

  In such an audio transmission device 1, the same sound is output in the entire facility (broadcasting), or the same sound is output in an appropriate area. In such a case, the voice receiving terminals 4 at a plurality of adjacent locations are required to have the same voice output timing. In order to meet such a demand, the audio transmission device 1 of the present embodiment is configured as follows.

  FIG. 1 is a block diagram showing the configuration of the audio transmission device 1. In FIG. 1, a voice transmission terminal 3 includes a voice input unit 11 for inputting analog voice and digital voice, a voice transmission unit 12 for transmitting the inputted voice as voice multicast data, and a voice before starting transmission of voice multicast data. And a learning packet transmitting unit 13 for transmitting a learning packet having a header configuration similar to that of a voice packet of multicast data. The voice input unit 11 corresponds to the sound source 5 in FIG.

  On the other hand, the voice receiving terminal 4 has received the voice receiving unit 14 that receives the voice multicast data, the learning packet discarding unit 15 that receives and discards the learning packet sent before starting the reception of the voice multicast data, A reception data determination unit 16 that determines whether the multicast data is received by the voice reception unit 13 or the learning packet discard unit 14, and a voice reproduced from the voice multicast data received by the voice reception unit 13 is output. And an audio output unit 17. The audio output unit 17 is composed of a speaker.

  About the audio | voice transmission apparatus comprised as mentioned above, the operation | movement is demonstrated using FIGS. 1-4. FIG. 3 shows an operation at the start of transmission in the voice transmitting terminal 3. When the voice transmission terminal 3 receives an instruction to start transmission by operating the operation unit (not shown) of the voice transmission terminal 3 (S21), first, the learning packet transmission unit 13 transmits a learning packet (S22). Then, the voice transmission unit 12 transmits a voice packet generated from the voice input by the voice input unit 11 (S23).

  FIGS. 4A to 4C show examples of voice packets and learning packets. FIG. 4A shows a voice packet. The voice packet is an IP multicast packet, and has a MAC layer, an IP layer, a UDP layer, and an RTP layer as a header configuration of the IP multicast packet as shown in the figure. Yes. The voice packet has voice data in the payload portion (data portion other than the header).

  On the other hand, FIG. 4B shows a learning packet, and the learning packet has the same header structure as the voice IP multicast packet. However, the learning packet is an IP packet having only a header portion and does not have a payload portion. The learning packet can be said to be a packet obtained by deleting the voice data portion from the voice packet. Such learning packets are generated and transmitted by the learning packet transmitter 13. Learning packets are also transmitted by multicast in the same manner as voice packets. FIG. 4C shows another example of the learning packet. As shown in this example, the learning packet may have dummy data in the payload portion instead of the voice data.

  Next, the operation on the receiving side will be described. When the IP multicast packet is received by the voice receiving terminal 4, first, the IP multicast packet is supplied to the reception data determination unit 16. The reception data determination unit 16 determines whether the received IP multicast packet is a voice packet or a learning packet.

  The reception data determination unit 16 is configured to determine whether the input packet data is data immediately after the start of transmission as a determination criterion. Packet data immediately after the start of transmission is determined to be a learning packet, and subsequent packet data is determined to be a voice packet. The number of learning packets may be one.

  As another determination criterion, a dedicated flag may be set in the extended portion of the RTP layer in the packet header configuration. In the case of a learning packet, a flag is set. The received data determining unit 16 refers to the flag to determine whether the received packet is a learning packet or a voice packet. If the flag is set, the reception data determination unit 16 determines that the reception packet is a learning packet.

  A packet determined to be a learning packet by the reception data determination unit 16 is received by the learning packet discard unit 15. Then, the learning packet discard unit 15 performs processing for discarding the learning packet as invalid data.

  On the other hand, a packet determined as a voice packet by the reception data determination unit 16 is received by the voice reception unit 14. The voice receiver 14 receives the voice packet, acquires voice multicast data, and reproduces the voice data. An analog audio signal is generated from the audio multicast data in the audio receiving unit 14, supplied to the audio output unit 17, and audio is output from the audio output unit 17.

  The operation of the voice transmission device 1 has been described separately for the transmission side and the reception side. The audio transmission device 1 operates as follows as a whole. At the start of transmission, a learning packet is first transmitted from the voice transmitting terminal 3 to the IP network 2 by multicast and transmitted to a plurality of voice receiving terminals 4. With the transmission of the learning packet, learning is performed by components such as various switches such as a hub constituting the IP network 2, and as a result, the network path is determined in the IP network 2. Then, after the network path is determined by the learning packet, the voice packet is transmitted from the voice transmission terminal 3 to the voice reception terminal 4. At the time when transmission of voice packets starts, the network path is determined. A network component such as a hub normally continues the same operation, so that the audio packet of the stream continues to be transmitted along the same route as the learning packet. Since voice packets pass through a predetermined network path, voice reception is started almost simultaneously at the plurality of voice receiving terminals 4, and there is almost no difference in reception start time. Therefore, audio output from the plurality of audio receiving terminals 4 is started almost simultaneously.

  This embodiment is advantageous in the following points. In the present embodiment, since the output timing information is not incorporated in the audio data, the audio receiving terminal 4 outputs the received audio data immediately. In such a case, if the learning packet according to the present embodiment is not transmitted, the transmission of the voice packet is started without determining the network path, and the network path is determined in the process of transmitting the voice packet. Since the network route determination time varies between terminals, the reception start timing of voice packets also varies between terminals, and the voice output timing also varies. Variation in network route determination time is a major factor in variation in audio output timing. On the other hand, in the present embodiment, the network path is already determined at the time when the transmission of the voice packet is started by transmitting the learning packet in advance. Therefore, the variation factors as described above can be eliminated, the reception start timing of the voice packet can be made uniform, and the variation in the voice output timing can be greatly reduced.

  In this embodiment, a delay time from the transmission of the learning packet to the start of transmission of the voice packet is set in advance. This delay time is set according to the network route determination time. More specifically, the delay time is set longer than the standard network route determination time in the corresponding IP network 2 by a predetermined margin time. The delay time is, for example, several hundred msec.

  The audio transmission apparatus 1 according to the first embodiment of the present invention has been described above. According to the present embodiment, since the learning packet is transmitted before the voice packet, the network path of the IP network is determined by the transmission of the learning packet, and then the voice packet passes through the determined network path. Since the voice packet passes through the already determined network path, the time difference of the voice reception start time between the voice receiving terminals is reduced, and the fluctuation of the reception start time is reduced. Therefore, even when audio reproduction is performed at a plurality of locations, fluctuations in audio reproduction at a plurality of locations can be reduced. Even in a stream format that does not have output timing information in the stream, fluctuations in reception start and audio output can be minimized. And the shift | offset | difference of the sound level of the adjacent speaker in the receiving side can be suppressed, and an audio | voice output can be performed favorably.

  In the present embodiment, the learning packet for determining the network route has the same header configuration as the voice packet. This makes it possible to determine a network path regardless of the form of a switch such as a hub as a network component, which is also advantageous in this respect.

  And according to this Embodiment, since it is not necessary to incorporate output timing information in an audio | voice stream, it can avoid that an apparatus scale becomes large. Further, since the network route is determined with a simple configuration using the learning packet, the terminal device does not have to support a route deciding protocol such as IGMP. Therefore, the configuration of the terminal of the voice transmission device can be simplified. This point is particularly advantageous in the audio transmission apparatus 1 such as a large-scale facility to which the present embodiment is applied. In such an audio transmission device 1, the number of speakers is enormous, and may range from thousands to tens of thousands. Therefore, when shifting from a conventional analog system to a digital system, it is required to suppress the capability of each terminal device as much as possible. When each terminal device is required to have a high level of function, it is practically difficult to digitize the voice transmission device. Against this background, this embodiment is particularly advantageous because it can reduce fluctuations in voice output with a simple configuration in which learning packets are transmitted in advance.

  Next, an audio transmission apparatus according to the second embodiment of the present invention is shown in FIG. Hereinafter, description of matters common to the first embodiment is omitted.

  In FIG. 5, in the audio transmission device 101 of the present embodiment, the configuration of the audio transmission unit 12 is shown in more detail, and the audio transmission unit 12 includes a delay generation unit 121 and a transmission unit 122. The transmission unit 122 generates a voice packet and sends it to the IP network 2. The delay generation unit 121 generates a delay time from the learning packet transmission to the voice packet transmission.

  The delay generation unit 121 controls the delay time of the voice packet generation by the transmission unit 122. The transmission unit 122 is controlled by the delay generation unit 121 and starts transmitting voice packets after a delay time from the learning packet transmission by the learning packet transmission unit 13.

  The delay time generated by the delay generator 121 can be arbitrarily changed. The delay time is set as follows. When an IP multicast packet is sent to the IP network 2, the network route determination time varies depending on the configuration of the network and the difference in devices included in the network. Therefore, the route determination time in the target network is measured. Actually, an IP multi-packet may be transmitted and the route determination time may be measured. The delay time is set so as to be longer than the actual IP network 2 route determination time by a predetermined time. The set delay time is stored in the voice transmission terminal 3, and is referred to by the delay generation unit 121. The delay generation unit 121 generates the stored delay time.

  The setting of the delay time may be performed using an operation unit provided in the voice transmission terminal 3. Further, the route determination time may be measured by the voice transmission device 1, and the delay time may be automatically set according to the measurement result.

  Thus, in this embodiment, since the delay time is set based on the route determination time according to the actual configuration of the IP network 2, the delay time can be set to the minimum necessary time. Such a delay time setting is advantageous in the following points. If the delay time is too short, there is a possibility that transmission of voice packets will start before route determination. In this case, fluctuations in voice output timing may not be sufficiently reduced. On the other hand, if the delay time is long, the start of audio output is also delayed. In the present embodiment, the network route is determined in advance with a simple configuration using learning packets, and the network route is not reliably controlled. The route can be lost. In this embodiment, by controlling the delay time, the delay time can be set as small as possible within a range larger than the route determination time, and the fluctuation of the voice output is suitably reduced.

  The audio transmission device 1 according to the second embodiment of the present invention has been described above. According to the present embodiment, it is possible to specify the time interval (the above delay time) between the transmission of the learning packet and the start of transmission of the voice packet. Thereby, the start time of the voice packet can be appropriately set individually so that the transmission of the voice packet starts after the network path is determined by the transmission of the learning packet, and the fluctuation of the voice generation can be suitably reduced. Voice multicast data can be transmitted after the network route is determined regardless of the network system, and the difference in absolute reception time between the terminals on the receiving side can be minimized.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and it goes without saying that those skilled in the art can modify the above-described embodiments within the scope of the present invention.

  As described above, the audio transmission device according to the present invention can reduce fluctuations in the voice reception start time, and even when voice reproduction is performed at a plurality of places, the fluctuation of voice reproduction at a plurality of places is reduced. And is useful as an audio transmission device for private broadcasting.

The block diagram of the audio | voice transmission apparatus in the 1st Embodiment of this invention The figure which shows the whole constitution of the audio transmission equipment Flow chart showing operation of voice transmitting terminal (A) The figure which shows the packet structure of a voice packet (b) The figure which shows the packet structure of a learning packet (c) The figure which shows another example of the packet structure of a learning packet The block diagram of the audio | voice transmission apparatus in the 2nd Embodiment of this invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Voice transmission apparatus 2 IP network 3 Voice transmission terminal 4 Voice reception terminal 5 Sound source 6 Speaker 11 Voice input part 12 Voice transmission part 13 Learning packet transmission part 14 Voice reception part 15 Learning packet discard part 16 Received data judgment part 17 Voice output part

Claims (2)

A voice transmitting terminal for sending voice multicast data to an IP network; and a plurality of voice receiving terminals for receiving the voice multicast data transmitted from the voice sending terminal through the IP network;
The voice transmitting terminal multicasts a voice packet transmitting means for transmitting voice packets of the voice multicast data and a learning packet including a header structure similar to the voice packet before transmission of the voice packets is started. Learning packet transmission means,
Each of the plurality of voice receiving terminals includes a learning packet discarding unit that discards the learning packet received before the start of reception of the voice packet.   The voice transmission device according to claim 1, wherein a time interval between transmission of the learning packet by the learning packet transmission unit and transmission start of the voice packet by the voice transmission unit can be specified.

Images