CN101635723A - Method and device for transferring data frames - Google Patents
Method and device for transferring data frames Download PDFInfo
- Publication number
- CN101635723A CN101635723A CN 200910157626 CN200910157626A CN101635723A CN 101635723 A CN101635723 A CN 101635723A CN 200910157626 CN200910157626 CN 200910157626 CN 200910157626 A CN200910157626 A CN 200910157626A CN 101635723 A CN101635723 A CN 101635723A
- Authority
- CN
- China
- Prior art keywords
- frame
- formatted voice
- data
- voice frame
- speech
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000003044 adaptive effect Effects 0.000 claims description 60
- 238000006243 chemical reaction Methods 0.000 claims description 34
- 238000005538 encapsulation Methods 0.000 claims description 34
- 238000001228 spectrum Methods 0.000 claims description 33
- 239000012792 core layer Substances 0.000 claims description 32
- 238000013139 quantization Methods 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000011002 quantification Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 230000010076 replication Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 5
- 238000004806 packaging method and process Methods 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 15
- 239000012467 final product Substances 0.000 description 9
- 238000012856 packing Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013213 extrapolation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 239000010410 layer Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/16—Vocoder architecture
- G10L19/173—Transcoding, i.e. converting between two coded representations avoiding cascaded coding-decoding
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
- G10L19/16—Vocoder architecture
- G10L19/18—Vocoders using multiple modes
- G10L19/24—Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The embodiment of the invention discloses a method for transferring data frames, which comprises the following steps: receiving data frames with first format; and packaging payload data of at least one of two received continuous data frames into data frames with second format. The other method for transferring data frames comprises the following steps: receiving the data frames with second format; obtaining center layer data in the data frames with second format; and respectively packaging into two data frames with first format by using the center layer data of the data frames with second format as payload data. The embodiment of the invention also provides a device for transferring the data frames. The embodiment of the invention enables code streams with two formats to be directly mutually transferred without complex network negotiation process, and improves the reliability and the stability of transferring the data frames.
Description
Technical field
The present invention relates to communication technical field, be specifically related to the conversion method and the device of Frame.
Background technology
G.729 be a narrowband speech compression scheme that is widely used in VOIP communication, code rate is 8kb/s, and every 10ms is a frame.And G.729 G.729 appendix B is adding silence compression scheme on the basis, supports discontinuous transmission, can further save communication bandwidth.
G.729.1 be the encoding and decoding speech standard of a new generation, has the extendible framework of a kind of layering, G.729 being kernel, support 8kb/s to amount to 12 kinds of different encoding and decoding speed to 32kb/s, frame length 20ms, for with the 10ms frame difference of G.729 (B), the frame of frame length 20ms is called superframe in this article.
G.729.1C be a kind of new G.729.1 silence compression scheme, frame length 20ms has the layering extensible architecture equally, and G.729B to serve as basic layer, maximum SID frame bit length is 43 bits.
Three kinds of frame types are arranged G.729B: speech frame (SP, Speech), effectively payload is 80 bits; Silence description frames (SID, Silence Insertion Descriptor), effectively payload is 15 bits; (NT, NO_DATA), effectively payload is 0 to the tone-off frame.Three kinds of frame types are arranged G.729.1C: speech frame (SP), effectively payload does not wait from 160 bit to 640 bits; Silence description frames (SID), effectively payload does not wait from 15 bit to 43 bits; Tone-off frame (NT), effectively payload is 0.Because G.729B frame length is 10ms, and frame length G.729.1C is 20ms, and is G.729B different with frame length G.729.1C, so between may need code stream is repacked when carrying out intercommunication.
At present, in the encoder G.729.1C that prior art provides, support one to be specifically designed to the G.729B pattern of code stream " G729B_BST " of encoding, making G.729.1C can compatible code stream G.729B.But, G.729.1C need special order input indication and could start, and after the startup with G.729B compatible pattern, G.729.1C encoder reality is according to G.729B coded system, every 10ms frame of encoding out; G.729.1C decoder be input as G.729B code stream.This working method, in actual communication systems, at all can't operate as normal in the time of may causing G.729B with intercommunication G.729.1C.A simplification as shown in Figure 1 G.729B with the system of intercommunication G.729.1C:
In the terminal 1 of using encoder G.729B with use when carrying out intercommunication between the terminal 3 of decoder G.729.1C, at first, the gateway terminal 3 of need giving advance notice, terminal 1 will send code stream G.729B, therefore gateway can send an order, start the G729B_BST pattern of terminal 3, both sides begin intercommunication then.Similarly; use the terminal 2 of encoder G.729.1C and use when carrying out intercommunication between the terminal 4 of decoder G.729B; at first gateway need be notified terminal 2; terminal 4 code stream G.729B of can only decoding; therefore gateway need send an order; start the G729B_BST pattern of terminal 2, both sides could normal intercommunication then.Well imagine, lose owing to the reason of channel or make mistakes that although set up the passage of intercommunication like this between terminal 1 and the terminal 3, terminal 3 can not be correctly decoded the code stream of terminal 1 generation if gateway sends to order that terminal 3 starts the G729B_BST patterns; Similarly, if gateway sends to the order of terminal 2 startup G729B_BST patterns also intercommunication correctly between terminal 2 and the terminal 4 has taken place to lose or make mistakes.
Therefore, prior art has tangible unsteadiness about G.729B being easier to make mistakes with the method for intercommunication G.729.1C.
Summary of the invention
In view of this, the invention provides a kind of conversion method and device of Frame, can improve the stability of Frame conversion.
The conversion method of a kind of Frame that the embodiment of the invention provides comprises:
Receive first kind of format data frame;
One of at least payload data is packaged into second kind of format data frame in continuous two described first kind format data frame that utilization receives.
The conversion method of the another kind of Frame that the embodiment of the invention provides comprises:
Receive second kind of format data frame;
Obtain the core layer data in described second kind of format data frame;
Utilize described core layer data to be packaged into two first kind of format data frame respectively as payload data.
The embodiment of the invention also provides a kind of Frame conversion equipment, comprising:
Receiving element is used to receive first kind of format data frame;
Encapsulation unit is used for utilizing continuous two described first kind format data frame payload data one of at least to be packaged into second kind of format data frame.
The another kind of Frame conversion equipment that the embodiment of the invention provides comprises:
Receiving element is used to receive second kind of format data frame;
Data extracting unit is used for extracting the core layer data of described second kind of format data frame;
Encapsulation unit is used for described core layer data are packaged into two first kind of format data frame respectively as payload data.
In the technical scheme that the embodiment of the invention provides, by extracting a kind of payload data of format data frame; And utilize the payload data that is extracted to be packaged into another kind of format data frame, and make the code stream of two kinds of forms directly to change mutually, and do not need complicated network negotiate process, improved the reliability and stability of Frame conversion.
Description of drawings
Fig. 1 be realization of the prior art G.729B with G.729.1C the conversion the system architecture schematic diagram;
Fig. 2 in the embodiment of the invention G.729B Frame be converted to G.729.1C Frame schematic diagram;
Fig. 3 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 4 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 5 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 6 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 7 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 8 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Fig. 9 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Figure 10 is according to the structural representation of the G.729.1C Frame of the structure of Frame G.729B in the embodiment of the invention;
Figure 11 is the neighbouring relations schematic diagram of two continuous G.729B Frames in the embodiment of the invention;
Figure 12 is according to the structural representation of the G.729.1C Frame of the structure of Frame G.729B in the embodiment of the invention;
Figure 13 in the embodiment of the invention G.729.1C speech frame be split into 2 G.729 schematic diagrams of speech frame;
Figure 14 in another embodiment of the present invention G.729.1C speech frame be split into 2 G.729 schematic diagrams of speech frame;
Figure 15 a constitutes schematic diagram for the conversion equipment of the Frame that provides in the embodiment of the invention;
Figure 15 b constitutes schematic diagram for another kind of Frame conversion equipment is provided in the embodiment of the invention;
Figure 16 constitutes schematic diagram for the conversion equipment of the Frame that provides in another embodiment of the present invention.
Embodiment
The embodiment of the invention provides a kind of data frame format conversion method and device, rebuild Frame when between the Frame of different-format, carrying out intercommunication, making gateway not need to send special order to opposite equip. goes indication to start the relevant work pattern, guaranteed the stability of communication system, the live load of gateway when having reduced the intercommunication operation greatly.Need to prove that the Frame in the embodiment of the invention specifically can comprise: speech frame, silence description frames (abbreviating quiet frame as), tone-off frame.Wherein, quiet frame, tone-off frame also can synthesize non-speech frame.
The conversion method of a kind of Frame that the embodiment of the invention provides is second kind of form from first kind of format conversion, comprises the steps:
Receive first kind of format data frame;
Utilize the payload data of continuous two described first kind of format data frame to be packaged into second kind of format data frame.Different situations for continuous two first kind of format data frame have following different disposal method:
One, for continuous two first kind of situation that format data frame is first kind of formatted voice frame, being packaged into second kind of format data frame comprises: the payload data and the former frame payload data of back one frame are merged, and the payload data after will merging is packaged into second kind of formatted voice frame, and the payload length value of second kind of formatted voice frame is the payload length value sum of described two first kind of formatted voice frames;
Two, the situation (as first kind of formatted voice frame in preceding, first kind of form non-speech frame in back or opposite order) that comprises first kind of formatted voice frame and first kind of form non-speech frame for continuous two first kind of format data frame, be packaged into second kind of format data frame can for: utilize first kind of formatted voice frame second kind of formatted voice frame of structure or abandon this first kind of formatted voice frame.Explanation respectively below:
1, utilizes second kind of formatted voice frame of first kind of formatted voice frame structure.Concrete building method can comprise:
1) constructs second kind of formatted voice frame by the payload that copies in described first kind of formatted voice frame; Or
2) parameter of the described first kind of formatted voice frame of extraction, described parameter one of comprises in line spectrum pair, adaptive codebook delay, adaptive codebook gain and the fixed codebook gain parameter at least, by the parameter of described first kind of formatted voice frame and the relevant parameter of speech frame before being carried out interpolation and re-quantization is constructed second kind of formatted voice frame.Specifically can comprise: decode parameter from this first kind of formatted voice frame payload, this parameter one of comprises in line spectrum pair, adaptive codebook delay, adaptive codebook gain and the fixed codebook gain parameter at least; The parameter and the relevant parameter of speech frame are before carried out parameter after interpolation obtains interpolation; Parameter after parameter after the interpolation quantized to obtain quantizing, utilizing the parameter after quantizing is one first kind form interpolation speech frame according to first kind of format combination; The payload of first kind of formatted voice frame and the payload data of described first kind of form interpolation speech frame are merged, and the payload data after will merging is packaged into second kind of formatted voice frame.
In addition, for the processing of first kind of form non-speech frame can for: 1) abandon this first kind of form non-speech frame; Or
2) if this non-speech frame is first kind of quiet frame of form, then this first kind of quiet frame of form of buffer memory utilizes one second kind quiet frame of form of payload data structure of described first kind of quiet frame of form to the next frame period.
2, abandon first kind of formatted voice frame, at this moment for the processing of first kind of form non-speech frame can for: when first kind of form non-speech frame is the quiet frame of first kind of form, utilize the quiet frame of this first kind of form to be packaged into the quiet frame of second kind of form;
Three, described continuous two first kind of format data frame comprise first kind of quiet frame of form and first kind of form tone-off frame condition (as first kind of quiet frame of form at preceding, first kind of form tone-off frame in back or opposite order), describedly are packaged into second kind of format data frame and comprise:
Utilize the quiet frame of this first kind of form to be packaged into the quiet frame of second kind of form.Be specifically as follows: 1, directly with the quiet frame of this first kind of form quiet frame as second kind of form; Or
2, utilize the payload of first kind of quiet frame of form to add that synchronization character and frame length value are packaged into one the second kind quiet frame of form.
The conversion method of the another kind of Frame that the embodiment of the invention provides is first kind of form from second kind of format conversion, comprising:
Receive second kind of format data frame;
Obtain the core layer data in this second kind of format data frame;
Utilize these core layer data to be packaged into two first kind of format data frame respectively as payload data.
Specifically comprise following three kinds of situations:
I, first kind of format data frame are first kind of quiet frame of form;
Then being packaged into two first kind of format data frame respectively is specially: the core layer data encapsulation in described second kind of quiet frame of form is become one the first kind quiet frame of form and one first kind form tone-off frame.
II, first kind of format data frame are first kind of formatted voice frame;
Then being packaged into two first kind of format data frame respectively is specially: utilize that the first core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame; Utilize second kind of formatted voice frame kind second core layer data encapsulation to become first kind of formatted voice frame.
III, first kind of format data frame are first kind of form tone-off frame, this means and do not receive any packet, therefore also do not need to send any packet.
Described first kind of form can be for G.729B, and described second kind of form is for G.729.1C.
Particularly, for example for speech coding frame format G.729B and G.729.1C, its conversion is divided into two classes, i.e. G.729B encoder and the G.729.1C conversion during intercommunication between the decoder, and G.729.1C encoder and G.729B between the decoder intercommunication the time conversion, below both of these case is described respectively.
One, G.729B encoder and G.729.1C intercommunication between the decoder
Because three kinds of frame types are arranged G.729B: speech frame (SP), effectively payload is 80 bits; Silence description frames (SID) abbreviates quiet frame as in the embodiment of the invention, effectively payload is 15 bits; Tone-off frame (NT), effectively payload is 0.Three kinds of frame types are also arranged G.729.1C: speech frame (SP), effectively payload does not wait from 160 bit to 640 bits; Silence description frames (SID), effectively payload does not wait from 15 bit to 43 bits; Tone-off frame (NT), effectively payload is 0.Because frame length G.729B is 10ms, and frame length G.729.1C is 20ms, therefore in actual communication systems, only need every 20ms in the time, with the packet G.729B of two 10ms receiving (for encoding code stream, a packet is a frame, down together) make up, be packaged into the G.729.1C packet of a 20ms.Shown in Figure 2, with 2 G.729B data envelope dress up a G.729.1C packet, with back 1 G.729B packet and preceding 1 G.729B payload data merging of packet, and the payload data after will merging is packaged into G.729.1C packet, and G.729.1C the payload length value of packet is two G.729B payload length value sums of packet.
But, if directly combination is packaged into G.729.1C code stream to code stream G.729B in twos, decoder may occur G.729.1C can't the recognition data packet length, for example, G.729B a SP frame and a SID frame is packaged into a superframe G.729.1C, then effectively payload is 95 bits, but in fact G.729.1C decoder be to discern the packet that payload is 95 bits.In this case, G.729.1C decoder still can't operate as normal.
Below with regard to adjacent two G.729B the frame situation that may occur describe.
1. speech frame is adjacent with quiet frame:
As previously mentioned, G.729B packet at two continuous 10ms among Fig. 3, the G.729B packet of these two 10ms directly can not be packaged into the G.729.1 packet of 20ms, and can only or be packaged into the G.729.1C speech frame of a 20ms this two frame, be packaged into the G.729.1C quiet frame of a 20ms, two kinds of possible solutions arranged:
1) abandon before the payload of 10msG.729B speech frame, only use 10ms quiet frame in back directly as the G.729.1C quiet frame of 20ms.Because interval minimum G.729B between the quiet frame is 20ms, that is to say that be quiet frame G.729.1C as if current 20ms, ensuing 20ms is the tone-off frame, as shown in Figure 4.
It should be noted that the tone-off frame does not need to send, that is to say the tone-off frame, be the notion in a time interval, and can not receive any information, also do not need to send again any information with respect to gateway.
2) the G.729B quiet frame of 10ms behind the buffer memory, the G.729.1C speech frame of a 20ms of the G.729B speech frame of 10ms structure sends before utilizing, the packet payload length is 160 bits, when treating next 20ms, the G.729.1C quiet frame that the G.729B quiet frame of the 10ms of buffer memory is configured to a 20ms sends, the packet payload length is 15 bits, as shown in Figure 5, the number of speech frames certificate of payload " SP ' " part wherein for obtaining by G.729B " SP " frame structure, building method can be direct copy, also can be the line spectrum pair in the speech frame G.729B, adaptive codebook postpones, parameters such as code book gain decode out, and relevant parameter with former frame carries out interpolation and re-quantization obtains.Need to prove, when reality is used, also can only carry out interpolation, and not need top described 4 kinds of parameters are all carried out interpolation, for example only line spectrum pairs parameter is carried out interpolation, and keep other the bit of parameter correspondence constant partial parameters.
In addition, the position of SP ' and SP is flexibly, promptly can be with the number of speech frames that constructs according to before putting, also can be with the number of speech frames that receives according to before putting, below among the embodiment position of SP ' and SP also be similar.
2.G.729B quiet frame is adjacent with the tone-off frame:
With reference to Fig. 6, when continuous two G.729B frame be quiet frame and tone-off frame, can directly the G.729B data envelope of two 10ms be dressed up the G.729.1C packet of a 20ms, the payload length after the encapsulation is 15 bits, be G.729.1C discernible long data packet, as shown in Figure 7.
3.G.729B the tone-off frame is adjacent with quiet frame:
When continuous two G.729B frame be tone-off frame and quiet frame, as shown in Figure 8.In this case, also can directly dress up the G.729B data envelope of two 10ms the G.729.1C packet of a 20ms, the payload length after the encapsulation is 15 bits, is G.729.1C discernible long data packet, as shown in Figure 7.
4.G.729B tone-off frame and speech frame are adjacent, as shown in Figure 9.
The G.729B packet of two 10ms directly can not be merged into the G.729.1C packet of a 20ms, because 80 bits are not the payload length that G.729.1C can discern.For this situation, has the packet G.729.1C that two kinds of solutions obtain 20ms.
1) data configuration in the speech frame of utilization back 10ms goes out the data of the speech frame of another one 10ms, as the data in the packet G.729.1C, the packet of two 10ms is formed the G.729.1C packet of 20ms, as shown in figure 10, wherein " SP ' " represent to construct the number of speech frames certificate that obtains by " SP " frame G.729B, building method can be direct copy, also can be parameters such as the line spectrum pair in the speech frame G.729B, adaptive codebook delay, code book gain to be decoded out relevant parameter with former frame carries out interpolation and re-quantization obtains.
2) abandon the speech frame of 10ms afterwards, current 20ms is handled as the tone-off frame, that is to say that gateway does not need to send any packet.
5. quiet frame is adjacent with speech frame
As shown in figure 11, similar with situation 1, for this situation two kinds of possible solutions are arranged also.
1) abandon the payload of 10ms speech frame afterwards, the payload of the quiet frame of 10ms as shown in Figure 2, utilizes the payload of the quiet frame of 10ms to add that synchronization character and frame length value are packaged into the G.729.1C quiet frame of a 20ms, as shown in figure 12 before only using.
2) abandon before the G.729B quiet frame of 10ms, utilize the G.729.1C speech frame of the 20ms of G.729B speech frame structure of back 10ms to send, the packet payload length is 160 bits, as shown in figure 10.Wherein " SP ' " represent to construct the number of speech frames certificate that obtains by " SP " frame G.729B, building method can be direct copy, also can be parameters such as the line spectrum pair in the speech frame G.729B, adaptive codebook delay, code book gain to be decoded out relevant parameter with former frame carries out interpolation and re-quantization obtains, the position of SP ' and SP is before promptly the number of speech frames certificate that constructs can being put flexibly, before also the number of speech frames certificate that receives can being put.
6. continuous two 10ms are speech frames G.729B, can directly encapsulate according to mode shown in Figure 2 to get final product.
Two, G.729.1C encoder and G.729B intercommunication between the decoder
Because code stream G.729.1C has only core layer just with G.729B compatible, therefore at encoder G.729.1C during with decoder intercommunication G.729B, in the embodiment of the invention, gateway only need abandon the data beyond the core layer in the packet G.729.1C, the packet that the packet of 20ms is divided into two 10ms gets final product then, according to the frame type difference of packet G.729.1C, be divided into following three kinds of situations:
If A. the G.729.1C packet of current 20ms is a speech frame, earlier the data beyond the core layer are abandoned, the G.729B packet that then payload of two 10ms frames is packaged into 10ms respectively gets final product, as shown in figure 13.
If B. the G.729.1C packet of current 20ms is quiet frame, need earlier the data except that core layer to be abandoned, the G.729B quiet frame that then core layer is packaged into 10ms sends, and another one 10ms is that the tone-off frame does not need to send, as shown in figure 14.
If C. G.729.1C current 20ms's is the tone-off frame, this means and do not receive any packet that therefore also do not need to send any packet, the G.729B decoder of decoding end is handled by the tone-off frame automatically.
This shows that the embodiment of the invention is by extracting a kind of payload data of format data frame; And utilize the payload data that is extracted to be packaged into another kind of format data frame, make the code stream of two kinds of forms directly to change mutually, and do not need complicated network negotiate process, realized the effective conversion of Frame, improved systematic function, as, reliability and high efficiency, the embodiment of the invention does not need to carry out conversion between the different-format by decode method that a kind of form is encoded to another form again of gateway in addition, therefore, can reduce the processing burden of gateway widely, save gateway resource.
Be described below in conjunction with some specific embodiments.For the purpose of understanding, G.729.1C following embodiment is converted to format data frame G.729B that format data frame is that example is described.
Embodiment one
In the present embodiment, transmitting terminal is encoder G.729B, receiving terminal is decoder G.729.1C, gateway receives the G.729B format data frame that transmitting terminal sends, as shown in table 1, need be converted to the G.729.1C Frame of form, promptly being packaged into frame length again is the G.729.1C Frame (as shown in table 2) of 20ms, exports to receiving terminal afterwards and is decoder G.729.1C.
Table 1:G.729B code stream
In the present embodiment, the G.729B code stream of input gateway is 10ms one frame, that is to say that transducer receives two frames Frame G.729B in 20ms, and the G.729.1C Frame that is encapsulated as 20ms again sends.Specific as follows:
101) when continuous two G.729B frame be respectively speech frame and quiet frame, referring to Fig. 3, abandon the number of speech frames certificate of previous 10ms, utilize the quiet frame data of a back 10ms to be packaged into the G.729.1C quiet frame data bag of a 20ms;
102) when continuous two G.729B frame be respectively quiet frame and tone-off frame, as Fig. 6 and shown in Figure 8, then directly be encapsulated as the G.729.1C quiet frame data bag of a 20ms;
103) when continuous two G.729B frame be respectively tone-off frame and speech frame, as shown in Figure 9, the number of speech frames that then directly abandons a back 10ms is according to getting final product, current 20ms is the tone-off frame, does not need to send any data;
104) when continuous two G.729B frame be respectively quiet frame and speech frame, as shown in figure 11, then abandon the number of speech frames certificate of a back 10ms, utilize the quiet frame data of previous 10ms, be encapsulated as the G.729.1C quiet frame data bag of a 20ms;
105) for two continuous frames be the situation of speech frame, can directly merge into a frame to two frames according to method shown in Figure 2 and get final product;
106) for two continuous frames be the situation of tone-off frame, need do not do any extra operation, also need not send any data.
According to above method, in the present embodiment, the output code flow of gateway is as shown in table 2:
Table 2: the G.729.1C Frame after the encapsulation
For the tone-off frame in the table 1, in fact gateway is not received any data, just takies several time slots; For the tone-off frame in the table 2, in fact gateway does not need to send any data; In the table 2, the payload data of n frame correspondence in (n) expression table 1 in the G.729.1C Frame payload section after the encapsulation.
Embodiment two
Present embodiment and embodiment one are basic identical, its difference is, for continuous two G.729B frame be speech frame and non-speech frame, it or not the number of speech frames certificate that abandons simply wherein, but cushion non-speech frame as the case may be, speech frame is wherein handled, and the concrete operations step of present embodiment is as follows:
201) when continuous two G.729B frame be respectively speech frame and quiet frame, referring to Fig. 3, the quiet frame data of a 10ms behind the buffer memory at first, number of speech frames certificate with previous 10ms copies a back 10ms to again, the number of speech frames certificate of two 10ms is merged into the G.729.1C speech frame packet of a 20ms according to method as shown in Figure 2, when next 20ms, the quiet frame data of buffer memory are packaged as the G.729.1C quiet frame data bag of a 20ms;
202) when continuous two G.729B frame be quiet frame and tone-off frame, as Fig. 6 and shown in Figure 8, be bagged directly into the G.729.1C quiet frame data bag of a 20ms;
203) when continuous two G.729B frame be tone-off frame and speech frame, as shown in Figure 9, the number of speech frames certificate of a back 10ms is copied to previous 10ms, the number of speech frames certificate of two 10ms is merged into the G.729.1C speech frame packet of a 20ms according to method as shown in Figure 2;
204) when continuous two G.729B frame be quiet frame and speech frame, as shown in figure 11, abandon the quiet frame data of previous 10ms, the number of speech frames certificate of a back 10ms is copied to previous 10ms, and the number of speech frames certificate of two 10ms is merged into the G.729.1C speech frame packet of a 20ms according to method as shown in Figure 2;
205) for two continuous frames be the situation of speech frame, can directly two frames be merged into a frame according to method shown in Figure 2 and get final product;
206) for two continuous frames be the situation of tone-off frame, need do not do any extra operation, also need not send any data.
According to above method, in the present embodiment, the output code flow of gateway is as shown in table 3:
Table 3: the G.729.1C Frame after the packing
For the tone-off frame in the table 1, in fact gateway is not received any data; For the tone-off frame in the table 3, in fact gateway does not need to send any data; In the table 3, the payload data of n frame correspondence in (n) expression table 1 in the G.729.1C Frame payload section after the packing.
Embodiment three
Present embodiment and embodiment two are basic identical, its difference is, each the frame speech frame that receives, use decoder decode G.729B goes out line spectrum pair, adaptive codebook delay, adaptive codebook gain and fixed codebook gain, and these have in short-term parameter and buffer memory stably, but do not need to rebuild voice signal.For needing to carry out the zone (payload data district) of number of speech frames among the embodiment two according to copy, after carrying out interpolation with the up-to-date relevant parameter that receives parameter and previous frame buffer memory, parameter field also quantizes again, upgrade the pitch delay Parity Check Bits, to receive bit that all the other bits in the Frame and interpolation and re-quantization obtain G.729B Frame again as a 10ms, and with the number of speech frames that receives according to according to method shown in Figure 2, merge the G.729.1C Frame that obtains a 20ms.The concrete operations step of present embodiment is as follows:
301) when continuous two G.729B frame be speech frame and quiet frame, referring to Fig. 3, the quiet frame data of a 10ms behind the buffer memory at first.Then line spectrum pair, adaptive codebook delay, adaptive codebook gain and fixed codebook gain in the number of speech frames certificate of previous 10ms are decoded out, obtain current line spectrum pair lsf respectively
Cur, adaptive codebook postpones lag
Cur, adaptive codebook gain g
a CurAnd fixed codebook gain g
f CurThe speech frame parameters note that last time of buffer memory decodes is done: line spectrum pair lsf
Pre, adaptive codebook postpones lag
Pre, adaptive codebook gain g
a PreAnd fixed codebook gain g
a PreUtilize the method for interpolation again, this parameter that receives is expanded, the method for interpolation can have multiple, for example:
I) interpolation:
P
New=α P
Cur+ (1-α) P
Pre0≤α≤1 wherein.
Ii) extrapolation:
P
New=(1+ β) P
Cur-β P
Pre0≤β≤1 wherein.
Typically, α=0.5, β=0.5.
Symbol P can be any one of above four kinds of coding parameters, and simultaneously, the interpolation of these four kinds of parameters also can be selected different interpolation methods as required.These four coding parameters after the interpolation are remembered work respectively: line spectrum pair lsf
New, adaptive codebook postpones lag
New, adaptive codebook gain g
a NewAnd fixed codebook gain g
a NewThese parameter utilizations algorithm G.729 that interpolation is gone out quantizes, and utilize the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits, line spectrum pair, adaptive codebook delay, adaptive codebook gain and the corresponding bit of fixed codebook gain that the bit that again this is received fixed codebook indices in the speech frame and fixed codebook symbol correspondence and re-quantization obtain are the G.729B speech frame of the new 10ms of a frame according to format combination G.729B.
With reference to Fig. 2, the speech frame of two 10ms (interpolation obtain and receive) data are merged into the G.729.1C speech frame packet of a 20ms, simultaneously, utilize lsf
Cur, lag
Cur, g
a CurAnd g
f CurUpgrade speech frame parameters buffer area lsf
Pre, lag
Pre, g
a PreAnd g
a PreWhen next 20ms, the quiet frame data of buffer memory are packaged as the G.729.1C quiet frame data bag of a 20ms.
Need to prove, when utilizing that obtain and these two the 10ms speech frames that receive of interpolation to merge according to the method for Fig. 2, can dispose the position of the payload data of two frames as required at new G.729.1C Frame, the number of speech frames that promptly interpolation can be gone out is according to before putting, before also the number of speech frames certificate that receives can being put.
302) when continuous two G.729B frame be quiet frame and tone-off frame, as Fig. 6 and shown in Figure 8, be bagged directly into the G.729.1C quiet frame data bag of a 20ms;
303) when continuous two G.729B frame be tone-off frame and speech frame, as shown in Figure 9, utilize 301) described in method, the number of speech frames of at first utilizing a back 10ms is according to expanding the new number of speech frames certificate of a frame, number of speech frames certificate with two 10ms puts in order as required again, according to method as shown in Figure 2, with 2 G.729B packet merge into the G.729.1C speech frame packet of a 20ms;
304) when continuous two G.729B frame be quiet frame and speech frame, as shown in figure 11, at first abandon the quiet frame data of previous 10ms, utilize 301 again) described in method, the number of speech frames certificate of a back 10ms is expanded the new number of speech frames certificate of a frame, number of speech frames certificate with two 10ms puts in order as required again, according to method as shown in Figure 2, merges into the G.729.1C speech frame packet of a 20ms;
305) for two continuous frames be the situation of speech frame, can directly merge into a frame to two frames according to method shown in Figure 2 and get final product;
306) for two continuous frames be the situation of tone-off frame, need do not do any extra operation, also need not send any data.
According to above method, in the present embodiment, the output code flow of gateway is as shown in table 4:
Table 4: the G.729.1C Frame after the packing
For the tone-off frame in the table 1, in fact gateway is not received any data; For the tone-off frame in the table 4, in fact gateway does not need to send any data; In the table 4, the payload data of n frame correspondence in (n) expression table 1 in the G.729.1C Frame payload section after the packing; The speech frame that " expansion SP (n) payload " expression in the table 4 utilizes the data interpolating of n frame in the table 1 to go out.
Need to prove, when reality is used, also can only carry out interpolation, and not need top described 4 kinds of parameters are all carried out interpolation, for example only line spectrum pairs parameter is carried out interpolation, and keep other the bit of parameter correspondence constant partial parameters.
Embodiment four
Present embodiment is the further optimization of embodiment three.Be that with the difference of embodiment three parameter of the SP frame of expansion not exclusively is to be obtained by the up-to-date relevant parameter interpolation that receives parameter and previous frame buffer memory also.The concrete operations step of present embodiment is as follows:
401) when continuous two G.729B frame be respectively speech frame and quiet frame, referring to Fig. 3, abandon the number of speech frames certificate of previous 10ms, utilize the quiet frame data of a back 10ms to be packaged into the G.729.1C quiet frame data bag of a 20ms;
402) when continuous two G.729B frame be quiet frame and tone-off frame, as Fig. 6 and shown in Figure 8, be bagged directly into the G.729.1C quiet frame data bag of a 20ms;
403) when continuous two G.729B frame be tone-off frame and speech frame, as shown in Figure 9, with the number of speech frames of 10ms according in line spectrum pair, adaptive codebook postpone to decode out, obtain respectively: line spectrum pair lsf
Cur, adaptive codebook postpones lag
CurThe speech frame parameters note that last time of buffer memory decodes is done: line spectrum pair lsf
Pre, adaptive codebook postpones lag
PreUtilize the method for interpolation again, this parameter that receives is carried out the interpolation expansion, as follows:
lsf
new＝α·lsf
cur+(1-α)·lsf
pre
lag
new＝α·lag
cur+(1-α)·lag
pre
Wherein but the span of α is 0≤α≤1, α in the present embodiment=0.5.These two coding parameters after the interpolation are remembered work respectively: line spectrum pair lsf
New, adaptive codebook postpones lag
NewThese parameter utilizations algorithm G.729 that obtains after the interpolation is quantized, and utilize the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits.
For adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewThen no longer adopt the method expansion of interpolation, but respectively from adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewChoose the index value that makes that wherein decoding and rebuilding signal energy sudden change is minimum in the pairing quantification code book.Can reduce to greatest extent like this influence to reconstructed speech.Adaptive codebook gain g in the present embodiment
a NewAnd fixed codebook gain g
f NewThe corresponding equal value of code book index value is 20.
Then, line spectrum pair, adaptive codebook delay, adaptive codebook gain and the corresponding bit of fixed codebook gain that this bit that receives fixed codebook indices in the speech frame and fixed codebook symbol correspondence and re-quantization are obtained are the G.729B speech frame of the new 10ms of a frame according to format combination G.729B.With reference to Fig. 2, the speech frame of two 10ms (newly obtain and receive) data are merged into the G.729.1C speech frame packet of a 20ms.Particularly, the payload data of new G.729B speech frame and the payload data of the current G.729B speech frame that receives are merged, the payload data after will merging again is packaged into G.729.1C formatted voice frame.
404) when continuous two G.729B frame be quiet frame and speech frame, as shown in figure 11, at first abandon the quiet frame data of previous 10ms, utilize 403 again) described in method, the number of speech frames certificate of a back 10ms is expanded the new number of speech frames certificate of a frame, G.729B number of speech frames certificate with two 10ms puts in order as required again, according to method as shown in Figure 2, the payload data of the G.729B speech frame of two 10ms is merged, payload data after will merging again is packaged into G.729.1C formatted voice frame, obtains the G.729.1C speech frame packet of a 20ms;
405) for two continuous frames be the situation of speech frame, can directly merge into a frame to two frames according to method shown in Figure 2 and get final product;
406) for two continuous frames be the situation of tone-off frame, need do not do any extra operation, also need not send any data.
According to above method, in the present embodiment, the output code flow of gateway is as shown in table 5:
Table 5: the G.729.1C Frame after the packing
For the tone-off frame in the table 1, in fact gateway is not received any data; For the tone-off frame in the table 5, in fact gateway does not need to send any data; In the table 5, the payload data of n frame correspondence in (n) expression table 1 in the G.729.1C Frame payload section after the packing; The speech frame that " expansion SP (n) payload " expression in the table 5 utilizes the data interpolating of n frame in the table 1 to go out.
Embodiment five
In the present embodiment, transmitting terminal is encoder G.729.1C, receiving terminal is decoder G.729B, gateway receives the G.729.1C format data frame that transmitting terminal sends, as shown in table 1, need be converted to the G.729B Frame of form, the G.729.1C Frame (as shown in table 5) that is about to frame length and is 20ms is divided into the G.729B Frame of two 10ms, exports to the G.729B decoder of receiving terminal afterwards.
In the present embodiment, the code stream of input gateway is for G.729.1C, and is as shown in table 6:
Table 6: the G.729.1C Frame of input
In the present embodiment, the G.729.1C code stream of input is 20ms one frame, that is to say that transducer receives frame Frame G.729.1C in 20ms, is divided into the G.729B Frame of two 10ms again, and concrete handling process is as follows:
At first, what no matter receive is speech frame or quiet frame, the bit except the narrowband core layer is abandoned.
501),, Frame G.729.1C is split into the G.729 speech frame packet of two 10ms and sends get final product then according to method shown in Figure 13 if the G.729.1C Frame that receives is a speech frame;
502) if the G.729.1C Frame that receives is quiet frame, then according to method shown in Figure 14, G.729.1C quiet frame is split into the G.729B quiet frame data bag of a 10ms and sends (if former frame is a speech frame, then needs to send at once the G.729B quiet frame of current 10ms; If G.729B preceding 10ms's is that quiet frame also can be postponed till next 10ms and sends, current 10ms then handles according to the tone-off frame), another one 10ms frame is the tone-off frame, reality does not need to send;
For two continuous frames is the situation of tone-off frame, need not do any extra operation, also need not send any data.
According to above method, in the present embodiment, the output code flow of gateway is as shown in table 7: for the tone-off frame in the table 5, in fact gateway is not received any data; For the tone-off frame in the table 7, in fact gateway does not need to send any data; In the table 7, the corresponding Frame payload of n frame (n=1 or 2) in the m superframe in SP (m.n) the expression table 6 in the G.729B Frame payload section after the packing.
Table 7:G.729B code stream
Embodiment six
A kind of Frame conversion equipment 1500 that the embodiment of the invention provides, this device can be gateway or network-switching equipment, this conversion equipment comprises:
Receiving element 1510 is used to receive first kind of format data frame;
Encapsulation unit 1520 is used to utilize the payload data of continuous two described first kind of format data frame to be packaged into second kind of format data frame.
Further, shown in Figure 15 a, encapsulation unit 1520 can comprise in merge cells 1521, structural unit 1522, discarding unit 1523 and the quiet encapsulation unit 1524 at least one, wherein:
Merge cells 1521 is used for when continuous two first kind form logarithmic data frame is first kind of formatted voice frame, and with the payload data and the merging of former frame payload data of back one frame, and the payload data after will merging is packaged into second kind of formatted voice frame;
Structural unit 1522 is used for when continuous two first kind form logarithmic data frame comprises first kind of formatted voice frame and first kind of form non-speech frame, utilizes second kind of formatted voice frame of first kind of formatted voice frame structure;
Discarding unit 1523 is used for abandoning this first kind of formatted voice frame when continuous two first kind of format data frame comprise first kind of formatted voice frame and first kind of form non-speech frame;
Quiet encapsulation unit 1524 is used for when continuous two first kind of format data frame comprise first kind of quiet frame of form, utilizes the quiet frame of this first kind of form to be packaged into the quiet frame of second kind of form.
Structural unit 1522 further can comprise replication module 1522a and/or (extraction module 1522b and interpolating module 1522c), and is not shown, wherein:
Replication module 1522a, the payload that is used for copying this first kind of formatted voice frame is constructed second kind of formatted voice frame;
Extraction module 1522b is used to extract the parameter of this first kind of formatted voice frame, and described parameter one of comprises in line spectrum pair, adaptive codebook delay, adaptive codebook gain and the fixed codebook gain parameter at least;
Interpolating module 1522c is used for that the parameter of this first kind of formatted voice frame and the relevant parameter of speech frame are before carried out interpolation and re-quantization is constructed second kind of formatted voice frame.
In addition, with reference to Figure 15 b, Frame conversion equipment 1500 also comprises:
Decoding unit 1531 is used for decoding parameter from the speech frame payload of continuous two first kind of form non-speech frame and speech frame, and described parameter comprises that line spectrum pair, adaptive codebook postpone;
Processing unit 1532 is used for that the described parameter and the relevant parameter of speech frame are before carried out line spectrum pair, adaptive codebook after interpolation obtains interpolation and postpones and quantize, and utilizes the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits;
Selected cell 1533 is used for respectively from adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewChoose the index value that makes that wherein decoding and rebuilding signal energy sudden change is minimum in the pairing quantification code book;
Speech frame construction unit 1534, be used for fixed codebook indices and the bit of fixed codebook symbol correspondence and line spectrum pair, adaptive codebook delay and selected adaptive codebook gain and the fixed codebook gain corresponding index value bit that re-quantization obtains, make up first kind of new formatted voice frame according to first kind of form with the current first kind of formatted voice frame that receives;
Encapsulation unit 1520 merges the payload data of new first kind of formatted voice frame of described speech frame construction unit structure and the payload data of the current first kind of formatted voice frame that receives, and the payload data after will merging is packaged into second kind of formatted voice frame.
If not speech frame is quiet frame, this Frame conversion equipment 1500 also comprises:
Discarding unit 1535 is used for described quiet frame data are abandoned.
First kind of form can be for G.729B in the present embodiment, and second kind of form can be for G.729.1C.
Embodiment seven
The another kind of Frame conversion equipment 1600 that the embodiment of the invention provides comprises:
Receiving element 1610 is used to receive second kind of format data frame;
Further, encapsulation unit 1630 can comprise first package module 1631, second package module 1632 and the 3rd package module 1633 at least one, wherein:
First package module 1631 is used for when this second kind of format data frame is second kind of quiet frame of form, and the core layer data encapsulation in second kind of quiet frame of form is become one the first kind quiet frame of form and one first kind form tone-off frame;
The 3rd package module 1633 when being used for this first kind of format data frame and being first kind of formatted voice frame, utilizes that the second core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame.
In sum, in the technical scheme that the embodiment of the invention provides, by extracting a kind of payload data of format data frame; And utilize the payload data that is extracted to be packaged into another kind of format data frame, make the code stream of two kinds of forms directly to change mutually, and do not need complicated network negotiate process, realized the effective conversion of Frame, improved systematic function, as, reliability and high efficiency, the embodiment of the invention does not need to carry out conversion between the different-format by decode method that a kind of form is encoded to another form again of gateway in addition, therefore, can reduce the processing burden of gateway widely, save gateway resource.
Obviously, those skilled in the art should be understood that, above-mentioned each unit of the present invention or each step can realize with the general calculation device, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation element forms, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in the storage device and carry out by calculation element, perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of unit in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.All any modifications of being done within the spirit and principles in the present invention, be equal to replacement, improvement etc., all be included in protection scope of the present invention.
Claims (28)
1, a kind of conversion method of Frame is characterized in that, comprising:
Receive first kind of format data frame;
One of at least payload data is packaged into second kind of format data frame in continuous two described first kind format data frame that utilization receives.
2, the method for claim 1 is characterized in that, when described continuous two first kind of format data frame are first kind of formatted voice frame, describedly are packaged into second kind of format data frame and comprise:
With payload data and the former frame payload data merging of back one frame, and the payload data after will merging is packaged into second kind of formatted voice frame, and the payload length value of second kind of formatted voice frame is the payload length value sum of described two first kind of formatted voice frames.
3, the method for claim 1 is characterized in that, when described continuous two first kind of format data frame comprise first kind of formatted voice frame and first kind of form non-speech frame, describedly is packaged into second kind of format data frame and comprises:
Utilize second kind of formatted voice frame of first kind of formatted voice frame structure.
4, method as claimed in claim 3 is characterized in that, the described second kind of formatted voice frame of first kind of formatted voice frame structure that utilizes comprises:
Construct second kind of formatted voice frame by the payload that copies in described first kind of formatted voice frame; Or
Extract the parameter of described first kind of formatted voice frame, by the parameter of described first kind of formatted voice frame and the relevant parameter of speech frame before being carried out interpolation and re-quantization is constructed second kind of formatted voice frame.
5, method as claimed in claim 4, it is characterized in that, the parameter of the described first kind of formatted voice frame of described extraction, construct second kind of formatted voice frame and specifically comprise by the parameter of described first kind of formatted voice frame and the relevant parameter of speech frame before being carried out interpolation and re-quantization:
Decode parameter from described first kind of formatted voice frame payload, described parameter one of comprises in line spectrum pair, adaptive codebook delay, adaptive codebook gain and the fixed codebook gain parameter at least;
The described parameter and the relevant parameter of speech frame are before carried out parameter after interpolation obtains interpolation;
Parameter after parameter after the described interpolation quantized to obtain quantizing, utilizing the parameter after the described quantification is one first kind form interpolation speech frame according to first kind of format combination;
The payload of described first kind of formatted voice frame and the payload data of described first kind of form interpolation speech frame are merged, and the payload data after will merging is packaged into second kind of formatted voice frame.
6, method as claimed in claim 3 is characterized in that, the described second kind of formatted voice frame of first kind of formatted voice frame structure that utilizes comprises:
When two continuous first kind of format data frame are tone-off frame and speech frame, from described first kind of formatted voice frame payload, decode parameter, described parameter comprises that line spectrum pair, adaptive codebook postpone;
The described parameter and the relevant parameter of speech frame are before carried out line spectrum pair, adaptive codebook after interpolation obtains interpolation postpone and quantize, and utilize the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits;
Respectively from adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewChoose the index value that makes that wherein decoding and rebuilding signal energy sudden change is minimum in the pairing quantification code book;
The bit of fixed codebook indices in the current first kind of formatted voice frame that receives and fixed codebook symbol correspondence and line spectrum pair, the adaptive codebook that re-quantization obtains are postponed and selected adaptive codebook gain and fixed codebook gain corresponding index value bit, make up first kind of new formatted voice frame according to first kind of form;
The payload data of described first kind of formatted voice frame newly and the payload data of the current first kind of formatted voice frame that receives are merged, and the payload data after will merging is packaged into second kind of formatted voice frame.
7, method as claimed in claim 3 is characterized in that, the described second kind of formatted voice frame of first kind of formatted voice frame structure that utilizes comprises:
When two continuous first kind of format data frame are quiet frame and speech frame, abandon described quiet frame data, from described first kind of formatted voice frame payload, decode parameter, described parameter comprises that line spectrum pair, adaptive codebook postpone;
The described parameter and the relevant parameter of speech frame are before carried out line spectrum pair, adaptive codebook after interpolation obtains interpolation postpone and quantize, and utilize the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits;
Respectively from adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewChoose the index value that makes that wherein decoding and rebuilding signal energy sudden change is minimum in the pairing quantification code book;
The bit of fixed codebook indices in the current first kind of formatted voice frame that receives and fixed codebook symbol correspondence and line spectrum pair, the adaptive codebook that re-quantization obtains are postponed and selected adaptive codebook gain and fixed codebook gain corresponding index value bit, make up first kind of new formatted voice frame according to first kind of form;
The payload data of described first kind of formatted voice frame newly and the payload data of the current first kind of formatted voice frame that receives are merged, and the payload data after will merging is packaged into second kind of formatted voice frame.
8, method as claimed in claim 4 is characterized in that, describedly is packaged into second kind of format data frame and comprises: abandon described first kind of form non-speech frame.
9, method as claimed in claim 4 is characterized in that, described non-speech frame is first kind of quiet frame of form, describedly is packaged into second kind of format data frame and comprises:
The described first kind of quiet frame of form of buffer memory utilizes one second kind quiet frame of form of payload data structure of described first kind of quiet frame of form to the next frame period.
10, the method for claim 1 is characterized in that, when described continuous two first kind of format data frame comprise first kind of formatted voice frame and first kind of form non-speech frame, describedly is packaged into second kind of format data frame and comprises:
Abandon described first kind of formatted voice frame.
11, method as claimed in claim 10 is characterized in that,
Described first kind of form non-speech frame is the quiet frame of first kind of form; Describedly be packaged into second kind of format data frame and comprise: utilize the quiet frame of described first kind of form to be packaged into the quiet frame of second kind of form.
12, the method for claim 1 is characterized in that, when described continuous two first kind of format data frame comprise first kind of quiet frame of form and first kind of form tone-off frame, describedly is packaged into second kind of format data frame and comprises:
Utilize the quiet frame of described first kind of form to be packaged into the quiet frame of second kind of form.
13, as each described method in the claim 1 to 12, it is characterized in that,
Described first kind of form is form G.729B, and described second kind of form is form G.729.1C.
14, a kind of conversion method of Frame is characterized in that, comprising:
Receive second kind of format data frame;
Obtain the core layer data in described second kind of format data frame;
Utilize described core layer data to be packaged into two first kind of format data frame respectively as payload data.
15, method as claimed in claim 14 is characterized in that, described second kind of format data frame is second kind of quiet frame of form;
Describedly utilize described core layer data to be packaged into two first kind of format data frame respectively to comprise as payload data:
Core layer data encapsulation in described second kind of quiet frame of form is become one the first kind quiet frame of form and one first kind form tone-off frame.
16, method as claimed in claim 14 is characterized in that, described first kind of format data frame is first kind of formatted voice frame; Describedly utilize described core layer data to be packaged into two first kind of format data frame respectively to comprise as payload data:
Utilize that the first core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame;
Utilize that the second core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame.
As each described method in the claim 14 to 16, it is characterized in that 17, described first kind of form is form G.729B, described second kind of form is form G.729.1C.
18, a kind of Frame conversion equipment is characterized in that, comprising:
Receiving element is used to receive first kind of format data frame;
Encapsulation unit is used for utilizing continuous two described first kind format data frame payload data one of at least to be packaged into second kind of format data frame.
19, device as claimed in claim 18 is characterized in that, described encapsulation unit comprises:
Merge cells is used for when described continuous two first kind of format data frame are first kind of formatted voice frame, and with the payload data and the merging of former frame payload data of back one frame, and the payload data after will merging is packaged into second kind of formatted voice frame.
20, device as claimed in claim 18 is characterized in that, described encapsulation unit comprises:
Structural unit is used for when described continuous two first kind of format data frame comprise first kind of formatted voice frame and first kind of form non-speech frame, utilizes second kind of formatted voice frame of first kind of formatted voice frame structure.
21, device as claimed in claim 20 is characterized in that, described structural unit comprises:
Replication module, the payload that is used for copying described first kind of formatted voice frame is constructed second kind of formatted voice frame; And/or
Extraction module is used to extract the parameter of described first kind of formatted voice frame, and described parameter one of comprises in line spectrum pair, adaptive codebook delay, adaptive codebook gain and the fixed codebook gain parameter at least;
Interpolating module is used for that the parameter of described first kind of formatted voice frame and the relevant parameter of speech frame are before carried out interpolation and re-quantization is constructed second kind of formatted voice frame.
22, device as claimed in claim 18 is characterized in that, also comprises:
Discarding unit is used for abandoning described first kind of formatted voice frame when described continuous two first kind of format data frame comprise first kind of formatted voice frame and first kind of form non-speech frame.
23, device as claimed in claim 18 is characterized in that, described encapsulation unit comprises:
Quiet encapsulation unit is used for when continuous two first kind of format data frame comprise first kind of quiet frame of form, utilizes the quiet frame of described first kind of form to be packaged into the quiet frame of second kind of form.
24, device as claimed in claim 18 is characterized in that, also comprises:
Decoding unit is used for decoding parameter from the speech frame payload of continuous two first kind of form non-speech frame and speech frame, and described parameter comprises that line spectrum pair, adaptive codebook postpone;
Processing unit is used for that the described parameter and the relevant parameter of speech frame are before carried out line spectrum pair, adaptive codebook after interpolation obtains interpolation and postpones and quantize, and utilizes the adaptive codebook after quantizing to postpone to upgrade the pitch delay Parity Check Bits;
Selected cell is used for respectively from adaptive codebook gain g
a NewAnd fixed codebook gain g
f NewChoose the index value that makes that wherein decoding and rebuilding signal energy sudden change is minimum in the pairing quantification code book;
The speech frame construction unit, be used for fixed codebook indices and the bit of fixed codebook symbol correspondence and line spectrum pair, adaptive codebook delay and selected adaptive codebook gain and the fixed codebook gain corresponding index value bit that re-quantization obtains, make up first kind of new formatted voice frame according to first kind of form with the current first kind of formatted voice frame that receives;
Described encapsulation unit merges the payload data of new first kind of formatted voice frame of described speech frame construction unit structure and the payload data of the current first kind of formatted voice frame that receives, and the payload data after will merging is packaged into second kind of formatted voice frame.
25, device as claimed in claim 24 is characterized in that, described non-speech frame is quiet frame, and this device also comprises:
Discarding unit is used for described quiet frame data are abandoned.
26, a kind of Frame conversion equipment is characterized in that, comprising:
Receiving element is used to receive second kind of format data frame;
Data extracting unit is used for extracting the core layer data of described second kind of format data frame;
Encapsulation unit is used for described core layer data are packaged into two first kind of format data frame respectively as payload data.
27, device as claimed in claim 26 is characterized in that, described encapsulation unit comprises:
First package module is used for when described second kind of format data frame is second kind of quiet frame of form, and the core layer data encapsulation in described second kind of quiet frame of form is become one the first kind quiet frame of form and one first kind form tone-off frame.
28, device as claimed in claim 26 is characterized in that, described encapsulation unit comprises:
Second package module when being used for described first kind of format data frame and being first kind of formatted voice frame, utilizes that the first core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame;
The 3rd package module when being used for described first kind of format data frame and being first kind of formatted voice frame, utilizes that the second core layer data encapsulation becomes first kind of formatted voice frame in described second kind of formatted voice frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910157626 CN101635723A (en) | 2008-07-25 | 2009-07-21 | Method and device for transferring data frames |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810134755.5 | 2008-07-25 | ||
| CN200810134755 | 2008-07-25 | ||
| CN 200910157626 CN101635723A (en) | 2008-07-25 | 2009-07-21 | Method and device for transferring data frames |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101635723A true CN101635723A (en) | 2010-01-27 |
Family
ID=41570019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910157626 Pending CN101635723A (en) | 2008-07-25 | 2009-07-21 | Method and device for transferring data frames |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN101635723A (en) |
| WO (1) | WO2010009660A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102143185A (en) * | 2011-03-31 | 2011-08-03 | 北京经纬恒润科技有限公司 | Data transmission method and data transmission device |
| CN106375063A (en) * | 2016-08-30 | 2017-02-01 | 上海华为技术有限公司 | Data transmission method and equipment |
| CN110557226A (en) * | 2019-09-05 | 2019-12-10 | 北京云中融信网络科技有限公司 | Audio transmission method and device |
| CN113726634A (en) * | 2021-08-19 | 2021-11-30 | 宏图智能物流股份有限公司 | Voice transmission system and method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004064041A1 (en) * | 2003-01-09 | 2004-07-29 | Dilithium Networks Pty Limited | Method and apparatus for improved quality voice transcoding |
| CN1270549C (en) * | 2003-06-08 | 2006-08-16 | 华为技术有限公司 | Method for realizing interconnection of mobile communication network |
| CN101622711B (en) * | 2006-12-28 | 2012-07-18 | 杰恩邦德公司 | Methods and systems for silence insertion descriptor (sid) conversion |
-
2009
- 2009-07-17 WO PCT/CN2009/072802 patent/WO2010009660A1/en active Application Filing
- 2009-07-21 CN CN 200910157626 patent/CN101635723A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102143185A (en) * | 2011-03-31 | 2011-08-03 | 北京经纬恒润科技有限公司 | Data transmission method and data transmission device |
| CN102143185B (en) * | 2011-03-31 | 2015-11-25 | 北京经纬恒润科技有限公司 | Data transmission method and data transmission device |
| CN106375063A (en) * | 2016-08-30 | 2017-02-01 | 上海华为技术有限公司 | Data transmission method and equipment |
| CN110557226A (en) * | 2019-09-05 | 2019-12-10 | 北京云中融信网络科技有限公司 | Audio transmission method and device |
| CN113726634A (en) * | 2021-08-19 | 2021-11-30 | 宏图智能物流股份有限公司 | Voice transmission system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2010009660A1 (en) | 2010-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1225723C (en) | Noise suppression | |
| JPS63500697A (en) | Multiplex digital packet telephone communication system | |
| CN101964188B (en) | Voice signal coding and decoding methods, devices and systems | |
| CA2381827A1 (en) | Apparatus and method for encoding and decoding tfci in a mobile communication system | |
| CN104205210A (en) | Decoding device and method, audio signal processing device and method, and program | |
| JP2002027023A (en) | Device and method for wireless data transmitting/ receiving | |
| EP0920693A1 (en) | Method and apparatus for improving the voice quality of tandemed vocoders | |
| US20080129556A1 (en) | Two stage loss-less compressor for a clear channel over a packet network | |
| CN102571966A (en) | Network transmission method for large extensible markup language (XML) document | |
| CN101635723A (en) | Method and device for transferring data frames | |
| EP0138080A2 (en) | Communication system using variable length character code and coding method | |
| CN102480335A (en) | Method and system for transmitting business data | |
| CN105052040A (en) | System and method for multi-stream compression and decompression | |
| WO2007114290A1 (en) | Vector quantizing device, vector dequantizing device, vector quantizing method, and vector dequantizing method | |
| CN101345053A (en) | Speech signal processing device | |
| CN109861998A (en) | A kind of plug-in type dynamic analysis system and method based on Big Dipper short message agreement | |
| WO2001063772A1 (en) | Method and apparatus for optimized lossless compression using a plurality of coders | |
| CN101617361B (en) | Method and apparatus for rate reduction of coded voice traffic | |
| CN100581253C (en) | Method and apparatus for decoding bit stream | |
| CN108512553B (en) | Truncated regeneration code construction method for reducing bandwidth consumption | |
| CN101170590B (en) | A method, system and device for transmitting encoding stream under background noise | |
| CN107615810A (en) | Header compression system and method for online network code | |
| CN101335597B (en) | Decoding method and apparatus | |
| JP2001358690A (en) | Digital line multiplexing device | |
| CN107294878A (en) | Data transmission method, system and relevant device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20100127 |