CN105306420B - Method, device and server for realizing cyclic playback of text-to-speech services - Google Patents

Abstract

Embodiments of the present invention provide a method, device and server for realizing cyclic playback of text-to-speech services. The method includes: when the TTS server completes a TTS service to the text information by using the media channel of the media server, judging whether the number of times the TTS server completes the TTS service to the text information reaches the loop play of the text information The number of times NUM, to obtain the judgment result; when the judgment result is no, interact with the TTS server, so that the TTS server can use the media channel to complete another TTS service for the text information. The embodiment of the present invention improves the performance of the media server when performing the TTS service.

Description

Method, device and server for realizing cyclic playback of text-to-speech services

technical field

The invention relates to the communication field, in particular to a method, a device and a server for realizing cyclic playback of text-to-speech services.

Background technique

Media Server (Media Server, referred to as MS) is an independent device that provides dedicated media resource functions in the softswitch system, and is also an important device in the packet network. Media processing, including conversion of video and audio RTP data streams to video and audio files. At the same time, it is also responsible for receiving the user's DTMF input through the terminal, playing the guidance voice of the service, and displaying the dynamic guidance picture. Its SIP protocol and MSML/MOML capabilities enable it to complete the entire session process and interact with users under the control of the application server APP.

The Media Service Control Unit (MSCU) is an important unit in the media server. It mainly completes capability negotiation with other entities, provides management and maintenance of resources itself, and controls other service resource units to complete complex services.

The Media Storage Transmit Unit-audio (hereinafter referred to as MSTU) is a business resource unit in the media server, which completes the storage of massive audio data, including the audio file playback function. There is an external network port on the media storage unit, which can directly send and receive through the external network port on the unit.

The media processing unit (Media Resource Unit, MRU) mainly completes the functions of media codec conversion, number collection and conference audio mixing.

Today, media servers are widely used. It can be mainly summarized as audio and video playback, number collection and conference functions.

The function of Text To Speech (TTS for short) is to recognize the input text information, convert it into audio, and send it to the user. At present, in the field of telecommunications, the application of TTS is basically to configure a special TTS server, and specify the TTS through signaling to send audio to the client to complete a service.

Fig. 1 is a schematic structural diagram of a system for implementing a TTS loop play service according to related technologies. As shown in Figure 1, the workflow of the system includes the following steps:

Step S101: The terminal initiates a call to activate the service of the APP. APP initiates a business process to the media server;

Step S102: the APP requests the media server to complete N times of TTS services by sending N times of SIP signaling;

Step S103: the media server requests TTS resources from the TTS server through SIP signaling, and controls the TTS server to complete service functions through the MRCP protocol;

Step S104: the TTS server sends the media packet to the terminal through the media server, and the TTS server reports information such as identification of playing duration to the media server.

The above is the current typical networking and signaling control process. The TTS server is used as a peripheral device of the media server. When the APP requests a service, it only initiates to the media server, and the media server judges the service type. When the service type is TTS application, the media server initiates a request to the TTS server, applies for resources, and controls the behavior of the TTS server. The text is recognized as audio and sent to the media server.

The above process can complete multiple TTS services and realize the same text recognition loop playback. However, considering that the media server will apply for internal media resources (MSTU internal and external resources, MRU resources) every time it receives an INFO (TTS) service. Release, and the process is complicated, greatly increasing the failure rate, especially when calling with a large amount of data, it will seriously affect the performance of the media server.

Contents of the invention

In view of this, the purpose of the embodiments of the present invention is to provide a method, device and server for realizing cyclic playback of text-to-speech services, so as to reduce the complexity of the media server's processing of internal media resources in order to support TTS service cyclic playback.

In order to solve the above technical problems, the embodiments of the present invention provide the following solutions:

Embodiments of the present invention provide a method for realizing cyclic playback of TTS services from text to voice, for a media server, including:

When the TTS server uses the media channel of the media server to complete a TTS service to the text information, it is judged whether the number of times the TTS server completes the TTS service to the text information reaches the number of times NUM of the text information is played repeatedly, Obtain the judgment result;

When the judgment result is no, interact with the TTS server, so that the TTS server can use the media channel to complete another TTS service for the text information.

Preferably, the judging whether the number of times the TTS server completes the TTS service for the text information reaches the number NUM of loop playback times of the text information also includes:

Receiving a TTS service request message for the text information sent by the application server, where the TTS service request message carries the number of loop playback times NUM;

The loop playback times NUM is analyzed from the TTS service request message.

Preferably, it also includes:

When receiving the TTS service request message for the text information sent by the application server, open the media channel.

Preferably, it also includes:

When the determination result is yes, the media channel is closed, and the APP server is notified that the NUM times of loop playing of the text information is completed.

Preferably, the codec type corresponding to the media channel is determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server.

The embodiment of the present invention also provides a device for realizing cyclic playback of TTS services from text to voice, which is used for a media server, including:

A judging module, configured to judge whether the number of times the TTS server completes the TTS service for the text information reaches the number of times for the text information when the TTS server completes a TTS service for the text information by using the media channel of the media server. Loop playback times NUM, get the judgment result;

An interaction module, configured to interact with the TTS server when the judgment result is negative, so that the TTS server can use the media channel to complete another TTS service for the text information.

Preferably, it also includes:

The receiving module is used to receive the TTS service request for the text information sent by the application server before the number of times that the TTS server completes the TTS service for the text information reaches the number NUM of loop play of the text information message, the TTS service request message carries the number of times NUM of loop play;

The parsing module is used to determine whether the TTS server completes the TTS service for the text information before reaching the number NUM of loop playback of the text information, and parse the loop playback from the TTS service request message Number of times NUM.

Preferably, it also includes:

The opening module is configured to open the media channel when receiving the TTS service request message for the text information sent by the application server.

Preferably, it also includes:

The closing and notification module is used to close the media channel when the judgment result is yes, and notify the APP server that the NUM times of loop playing of the text information is completed.

Preferably, the codec type corresponding to the media channel is determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server.

The embodiment of the present invention also provides a server including the above-mentioned device for realizing the cyclic playback of the TTS service from text to voice.

It can be seen from the above that the embodiments of the present invention have at least the following beneficial effects:

Complete another TTS service by using the media channel used to complete a TTS service, thereby avoiding the closing and re-opening of the media channel, thereby reducing the number of establishment and release of internal resources of the media server and the corresponding signaling interaction , thereby reducing the pressure on the media server to process resources and signaling, and improving the performance of the media server when performing TTS services.

Description of drawings

Fig. 1 shows the schematic diagram of the general process structure of realizing the TTS loop play service according to related technologies;

Fig. 2 represents a kind of flow chart of the step that realizes the method for cyclic playing from text to voice TTS service that the embodiment of the present invention provides;

Fig. 3 shows the schematic diagram of the interactive structure of the media server and each module of the preferred implementation mode of the embodiment of the present invention;

Fig. 4 is a schematic diagram of the signaling sequence exchange between the media server and each module in a preferred implementation mode of the embodiment of the present invention;

Fig. 5 shows a structural block diagram of a device for implementing cyclic playback of text-to-speech TTS services provided by an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

Fig. 2 shows a kind of flow chart of the steps of realizing the method for cyclic playback from text to voice TTS service provided by the embodiment of the present invention. Including the following steps:

Step 201, when the TTS server uses the media channel of the media server to complete a TTS service to the text information, judge whether the number of times the TTS server completes the TTS service to the text information reaches the loop play of the text information Number of times NUM, get the judgment result;

Step 202, when the judgment result is no, interact with the TTS server, so that the TTS server can use the media channel to complete another TTS service for the text information.

The method is for a media server.

It can be seen that by using the media channel used to complete the TTS service once to complete another TTS service, the closing and re-opening of the media channel is avoided, thereby reducing the number of establishment and release of internal resources of the media server and the corresponding information. Command interaction, thereby reducing the pressure on the media server to process resources and signaling, and improving the performance of the media server when performing TTS services.

In the embodiment of the present invention, the judging whether the number of times the TTS server completes the TTS service for the text information reaches the number NUM of loop playback times of the text information may also include:

Receiving a TTS service request message for the text information sent by the application server, where the TTS service request message carries the number of loop playback times NUM;

The loop playback times NUM is analyzed from the TTS service request message.

Further, it may also include:

When receiving the TTS service request message for the text information sent by the application server, open the media channel.

In the embodiment of the present invention, may also include:

When the determination result is yes, the media channel is closed, and the APP server is notified that the NUM times of loop playing of the text information is completed.

In the embodiment of the present invention, the codec type corresponding to the media channel may be determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server.

In order to illustrate the embodiments of the present invention more clearly, preferred implementation modes of the embodiments of the present invention are provided below.

In view of the complex processing flow of the TTS business with multiple text recognition in the prior art, and the disadvantages of the impact on performance, this preferred implementation mode provides a method, device and system for realizing cyclic playback of TTS services from text to voice, so as to solve the problem of In the prior art, the media server handles the problems of high failure rate and low performance of the TTS loop playing service.

In order to achieve the above purpose, a method for realizing TTS loop playback service is provided, including:

The media server receives the access request from the application server APP, and determines the set of codec types supported by the media server;

The media server receives the TTS service request applied by the APP, and applies to the TTS server for service resources according to the TTS service type;

The media server parses the INFO (TTS) field to obtain the number N of loop playback times. When a TTS service is completed, the media server does not release the local resources, keeps the media link with the TTS server, and then conducts the next MRCP negotiation to identify the text information. And the number of loop requests is determined according to N, and finally the TTS server sends the N times of audio playback converted from the text recognition to the terminal through the media service (only needs to apply for a resource once).

In order to achieve the above purpose, there is also provided a system for realizing the TTS loop playback service from text to voice, and the realization device includes:

The first processing module is used to receive the access request from the application server APP, and determine the codec type set supported by the media server;

The second processing module is used to receive the TTS service request applied by the APP, and apply for TTS service resources according to the TTS service type, and determine the number of loop playback times;

The third processing module is configured to negotiate with the TTS server according to the codec type set to obtain the negotiated audio codec type, and send the media service data packet to the terminal server through the media server according to the audio codec type.

Fig. 3 is according to this preferred implementation mode, each module inside the media server and APP, TTS and terminal server exchange structural diagram, as shown in Fig. 3, media control unit MSCU, is used for sending session initiation protocol SIP, MRCP signaling to TTS Server, SIP negotiation to negotiate and specify the audio codec type that the media server matches with the TTS server, MRCP signaling interaction to control the TTS server to recognize text and play content; the voice center switching unit MRU is used to receive TTS server data packets, and The media service data packet is sent to the media storage transmission audio unit MSTU; wherein, the MSCU controls the MSTU to send the media service data packet to the terminal.

Figure 4 is a schematic diagram of the exchange sequence between the media server and each module, and the detailed workflow is as follows:

Steps S410 and S420APP sends INVITE signaling to the media server for media negotiation, the media server selects the codec type through its own capability set, and uses the MSTU external port address as the address for interacting with the terminal; the APP server sends an INFO request to the media server, The content in INFO is to apply for TTS service, the media server parses the field of loop playback in INFO as N, and saves all information;

In step S430, the media server negotiates with the TTS server, and controls the TTS server to convert text into speech. As shown in FIG. 4, the step S430 may specifically include the following steps:

In step S4301, the media control unit MSCU initiates SIP signaling to the TTS server to negotiate codec types. At this time, the audio codec capability set negotiated in the INVITE signaling is owned by the media server, that is, all codec types supported by the MRU;

Step S4302, the TTS server replies to the INVITE message 200 OK, and notifies the media server of the negotiated audio codec type;

Step S4303, the MSCU applies for the MSTU external port resources on the media server side required by the TTS service, the MRU1 transcoding resource, and the MRU2 transcoding resource; Receive data from the MRU internal port to the identification audio packet sent by the TTS server, and open the media channel on the media server side;

Step S4304, the MSCU sends a TCP/IP link establishment request to the TTS server;

In step S4305, the MSCU accepts the reply message of establishing a TCP/IP connection request issued by the TTS server;

Step S4306, the MSCU sends an MRCP request message to the TTS server, indicating the text information that the TTS server needs to identify;

Step S4307, the TTS server replies to the MRCP request message, notifies the media server that text recognition is in progress, and sends an audio packet to the MSTU external port, and the MRU sends the data packet forwarded from the MSTU external port via NAT to the terminal;

Step S4308, the TTS server notifies the MSCU that this text recognition is completed, and the MSCU notifies the TTS server to close this TCP link, and the MSCU module parses the saved INFO (TTS) loop times N to judge whether it is necessary to initiate an MRCP request to the TTS server again, If you need to continue to identify and play, repeat steps S4304 to S4308;

Step S4309, when N times of TTS services are completed, the MSCU sends a bye request message to the TTS server, notifying the TTS server to release the SIP data area corresponding to this TTS service;

Step S4310, the media server receives the reply message of bye from the TTS server, releases the SIP data area on the media server side, and the service is completed.

Step S440, the media server sends an info message to the APP, reporting information such as playing duration;

In step S450, the APP sends a BYE signaling to the media server to release resources.

In the above-mentioned embodiment, the media server completes N times of MRCP requests and recognizes text processing through one negotiation result with the TTS server, which obviously reduces the pressure on the media server to process resources and signaling, and greatly improves the performance of the media server when performing TTS services .

Fig. 5 shows a structural block diagram of a device for realizing cyclic playback of TTS services from text to voice provided by an embodiment of the present invention. With reference to Fig. 5, an embodiment of the present invention provides a device for realizing cyclic playback of TTS services from text to voice, including :

Judging module 501, configured to judge whether the number of TTS services completed by the TTS server for the text information reaches the text information when the TTS server completes a TTS service for the text information by using the media channel of the media server. The loop playback times NUM, get the judgment result;

The interaction module 502 is configured to interact with the TTS server when the judgment result is no, so that the TTS server can use the media channel to complete another TTS service for the text information.

The device is used for a media server.

It can be seen that by using the media channel used to complete the TTS service once to complete another TTS service, the closing and re-opening of the media channel is avoided, thereby reducing the number of establishment and release of internal resources of the media server and the corresponding information. Command interaction, thereby reducing the pressure on the media server to process resources and signaling, and improving the performance of the media server when performing TTS services.

In the embodiment of the present invention, may also include:

The receiving module is used to receive the TTS service request for the text information sent by the application server before the number of times that the TTS server completes the TTS service for the text information reaches the number NUM of loop play of the text information message, the TTS service request message carries the number of times NUM of loop play;

The parsing module is used to determine whether the TTS server completes the TTS service for the text information before reaching the number NUM of loop playback of the text information, and parse the loop playback from the TTS service request message Number of times NUM.

In the embodiment of the present invention, may also include:

The opening module is configured to open the media channel when receiving the TTS service request message for the text information sent by the application server.

In the embodiment of the present invention, may also include:

The closing and notification module is used to close the media channel when the judgment result is yes, and notify the APP server that the NUM times of loop playing of the text information is completed.

In the embodiment of the present invention, the codec type corresponding to the media channel may be determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server.

An embodiment of the present invention also provides a server, which includes the above-mentioned device for realizing the cyclic playback of the TTS service from text to voice. The server is for example: a media server.

The above is only the implementation of the embodiment of the present invention. It should be pointed out that for those skilled in the art, without departing from the principle of the embodiment of the present invention, some improvements and modifications can also be made. These improvements and Retouching should also be regarded as the scope of protection of the embodiments of the present invention.

Claims (11)

1. A method for realizing cyclic play from text to voice TTS service, used for media server, is characterized in that, comprising: When the TTS server uses the media channel of the media server to complete a TTS service to the text information, it is judged whether the number of times the TTS server completes the TTS service to the text information reaches the number of times NUM of the text information is played repeatedly, Obtain the judgment result; When the judgment result is no, interact with the TTS server, so that the TTS server can use the media channel to complete another TTS service for the text information. 2. The method according to claim 1, wherein said judging whether said TTS server completes the number of times of TTS service to said text information before reaching the loop play times NUM of said text information, further comprises: Receiving a TTS service request message for the text information sent by the application server, where the TTS service request message carries the number of loop playback times NUM; The loop playback times NUM is analyzed from the TTS service request message. 3. The method according to claim 2, further comprising: When receiving the TTS service request message for the text information sent by the application server, open the media channel. 4. The method according to claim 1, further comprising: When the determination result is yes, the media channel is closed, and the APP server is notified that the playing of the text information has completed the loop playing times NUM. 5. The method according to claim 1, wherein the codec type corresponding to the media channel is determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server. 6. A device that realizes cyclic playback of TTS services from text to voice, is used for a media server, and is characterized in that, comprising: A judging module, configured to judge whether the number of times the TTS server completes the TTS service for the text information reaches the number of times for the text information when the TTS server completes a TTS service for the text information by using the media channel of the media server. Loop playback times NUM, get the judgment result; An interaction module, configured to interact with the TTS server when the judgment result is negative, so that the TTS server can use the media channel to complete another TTS service for the text information. 7. The device according to claim 6, further comprising: The receiving module is used to receive the TTS service request for the text information sent by the application server before the number of times that the TTS server completes the TTS service for the text information reaches the number NUM of loop play of the text information message, the TTS service request message carries the number of times NUM of loop play; The parsing module is used to determine whether the TTS server completes the TTS service for the text information before reaching the number NUM of loop playback of the text information, and parse the loop playback from the TTS service request message Number of times NUM. 8. The device according to claim 7, further comprising: The opening module is configured to open the media channel when receiving the TTS service request message for the text information sent by the application server. 9. The device according to claim 6, further comprising: The closing and notification module is used for closing the media channel when the judgment result is yes, and notifying the APP server that the playing of the text information has completed the number of loop playing times NUM. 10. The device according to claim 6, wherein the codec type corresponding to the media channel is determined by the media server through negotiation with the TTS server according to the set of codec types supported by the media server. 11. A server, characterized in that it comprises the device for realizing cyclic playback of TTS services from text to voice according to any one of claims 6 to 10.