The transmission system, transmitter, mobile station and method for transmitting speech signals

 

(57) Abstract:

The invention relates to a transmitting system used in mobile radio stations for transmission of speech signals with the mobile station on the stationary. The technical result - ensuring the recovery of the optimal transmission parameters of speech signals after the resumption of the speech signal after a pause. The invention consists in that allows you not to turn off during pauses filters, and to direct the filter, run coded excitation vectors with appropriately processed speech signal. The result is a smoother, barely perceptible transition between a comfortable noise and speech signal when the latter resumes, artificial background noise, the so-called comfort noise, occurs during shutdown in the receiver. 4 C. and 4 h.p. f-crystals, 3 ill.

The present invention relates to a broadcasting system that contains a transmitter that includes a speech encoder having a storage device for storing signals excitation filter device for filtering signals of excitation, the tool of choice for mapping a signal derived from the speech signal with the output means of pauses in speech, and to at least partially disable speech coder when it detects a pause in speech, and a means for transmitting the optimal excitation signal to the receiver, which includes a speech decoder to recover the optimal excitation signal and the speech signal.

This system and method is widely known, for example, from Bishnu S7 Atal, Vladimir Cuperman, Allen Geneho "Advances in Speech Coding," more specifically on pages 69-79. In particular, it is widely used in mobile radio stations for transmission of speech signals with the mobile station on the stationary. The mobile station typically powered by batteries, because most of the energy consumed by the transmitter, it is known that the transmitter together with associated blocks off during pauses in speech to save energy and extend the useful life of the batteries. However, due to the high complexity of the device, the speech coder this speech encoder requires a considerable amount of energy, in particular due to the fact that for each block of speech and all excitation signals necessary to address all of the memory cells in a memory device, and also expressed the excitation vectors must be tucked filtering to determine the optimal excitation vector that gives, for example, the lowest energy of the difference signal at the stage of forming the difference.

In patent WO 93/13516 dedicated to the encoder, indicates that the speech encoder is switched off during pauses in speech, and continues generating only a few parameters, such as coefficients coding with linear prediction and autocorrelation coefficients for which the detector detects a pause in speech, and of which displays information regarding the transmitted background noise. It can be assumed simultaneously disable the filtering device as its output signals are not directly required during pauses in speech. However, when the speech signal is resumed, it takes some time to filter came on full power after switching off so that during the transition period, there is not the optimal parameters for the transmission of speech signals.

Therefore, the present invention is to offer a transmitting system of the type described in the opening paragraph, which provides significant energy savings during pauses in speech, and in which the optimal transmission parameters of speech signals is restored almost immediately after the resumption of the speech signal after a pause.

According to the present invention this is achieved by the use of a detector for the turn signal.

According to the obtained solution at shutdown picker off addressing, reading, and very expensive filtering is stored in the storage device of the excitation vectors, since the operation is addressing, reading, and very expensive filtering are schemes that require the greatest amount of computation, and stores only the function of the filtering device as this filter device requires a small amount of energy. Because when you turn off the addressing of the storage device to the second filter will cease to receive the input signal, the first filter receives the next signal, derived from the speech signal, that is the only vector excitation, because ideally, the input signals of the two devices are the same. When resuming the speech signal after a pause the first filter when using it again will also demonstrate a more smooth transition to the full speech coding.

With regard to the optimal parameters for the transmission of speech signals, it is known that the storage device comprises a first storage subdevice, which contains certain excitation vectors, and the second storage subdevice, motortalk during pauses in speech, but also the sum of the weighted vector excitation of the first storage subdevice and weighted vector excitation of the second storage subdevice being recorded in the second storage subdevice. The application of additional vectors provides obtaining nearly optimal excitation vectors, issuing a very small differential signal, i.e., small signal errors. This is particularly effective for areas vocalized speech, because the speech signal is almost periodic and almost never does not change dramatically. This applies also to the case of resumption of the speech signal after a pause in speech. Therefore, in order to have as values the excitation of the most recent values, and during pauses in speech, what are the most recent values can be used immediately after the resumption of the speech signal, according to a variant implementation of the method which is the subject of the invention, it is convenient to take during pauses in speech additional excitation vectors from the first input of the second filter and to record them in the second storage subdevice. In the second storage subdevice when resuming the speech signal to be additional primaline parameters for transmission of speech signals.

The device for implementing the method which is the subject of the present invention are described in dependent claims.

Next, the embodiments of the invention will be described with reference to the drawings, in which:

in Fig. 1 shows a transmitting system, which can be used in the invention

in Fig. 2 shows a block diagram of a speech encoder in the transmitting station; and

in Fig. 3 shows the structure of a storage device consisting of two subdevice.

In the transmission system shown in Fig. 1 produced by the microphone 1 speech signal is converted speech encoder 4 in the transmitter 2 in the encoded speech signal. The encoded speech signal is transmitted by the transmitter 2 to the receiver via the data transmission channel 3. The data channel can be, for example, a radio channel, two-wire copper wire or optical fiber. In the receiver 5, the encoded speech signal is converted by the decoder in the restored speech signal using the loudspeaker 7 is converted into an acoustic signal.

The speech encoder shown in Fig. 2, consists of a storage device 12, which is on channel 15 receives signals addressing and operated is cnyh the memory cells, periodically and sequentially controlled and read by the control circuit 14. Read the excitation vectors appear on channel 13 after the stage of weighting, which is not shown in detail here, the channel 13 is connected to the edge switch on two areas 28. This switch in two directions is, obviously, an electronic switch.

First, it is assumed that the switch 28 is in the lower position, so that the weighted and read the excitation vectors are received on the channel 13 to the input 29 of the first filter 16.

Digitized speech signal before encoding is applied to the input 11, which is connected to the filter 22. For clarity, not shown, the device for removal from the speech signal different parameters, in particular for removing coding coefficients from the linear prediction. These coefficients coding linear prediction served on the filter 22 (filter analysis coding linear prediction) which, as a result, generates a channel 23 so-called residual signal. Such residual signals also represent the excitation vectors stored in the storage device 12.

The residual signs also apply the same filter coefficients. The output signals of the filters 16 and 24 serves to link the formation of the difference of 18 forming the difference between the two signals, and this differential signal means the signal of error, since the difference signal is a measure of the difference between the speech signal at the input 11 and the speech signal recovered from the excitation vectors. This differential signal is supplied to the processing unit 20, forming the average energy of the signal error. This average energy is delivered through the channel 21 to the control circuit 14, which stores the address of a vector of excitation, which found the lowest value of the average energy. This address is transmitted to the receiving station as a parameter intended for the transmission of voice.

In addition, a detector 26, which perceives as the speech signal at the input 11, and a residual signal on channel 23 and based on this determines whether the input 11 of the real speech signal or at the moment there is a pause in speech, and to the input 11 receives only the background noise. If the detector 26 detects the presence of a pause, channel 27 is transmitted to the signal, which disables picker 10, formed by the control circuit if the first filter 16 will no longer receive excitation vectors, the signal on channel 27 actuates a switch on two areas 28, so that the input 29 of the filter 16 receives the residual signal on channel 23. This signal is essentially optimal excitation vector, issued each time a channel 13, so that each time only one vector excitation. If the input 11 is routed back to the speech signal and the items selector 10 are turned on again and switch on two areas 28 is returned to the lower position, the filter 16 again receives on channel 13 stored in the memory and measured excitation vectors, one of which must be selected optimal.

The input 29 of the filter 16 is connected with the information input of the storage device 12. As shown in Fig. 3, the storage device 12 is formed in fact the two subdevice 121 and 122, which performs the control circuit 14 of Fig. 1 through the respective inputs of the addressing 15a and 15b. Storage subdevice 121 in shared memory is set to read-only, which is the number of permanently stored vectors excitation. Storage subdevice 122, on the other hand, is a random-access memory, input 126 which comes with the ROM 125, determining the amount of the excitation vector from memory subdevice 121 multiplied by the first weighting coefficient to the multiplier 124, and the excitation vector from the second storage subdevice 122, multiplied by generally another factor in the following multiplier 123. The first storage subdevice 121 may also consist of a set of permanent storage devices, which are switched upon detection of vocal bezvikonnaja element in the speech signal.

Because the storage device 12 in Fig. 1 during pauses in speech, off, during this period of time on the channel 13 is not generated any excitation vectors. Therefore, the information input 126 of the second storage subdevice 122 in Fig. 3 directly connected to the input 29 of the filtering device 16, which also receives the signal during pauses in speech, i.e., the residual signal from the channel 23. Thus the second storage subdevice 122 contains the most recent vectors excitation and during breaks, so when moving speech signal simultaneously on channel 13 receives almost optimal excitation vectors.

1. Transmitting the speech signal containing pealow excitation, the first filter associated with the storage device to filter the excitation signals, the selector associated with the first filter and the second filter, for comparing the signal derived from the speech signal through a second filter, with the output signal of the first filter and to select the optimum excitation signal, a detector for detecting pauses in speech and to control the switch that turns off at least part of the speech coder when it detects a pause in speech, wherein the switch is configured to turn off the picker and the storage device in case of pauses in speech, off of the first filter from the storage device and connected to the first filter further signal derived from the speech signal.

2. The transmission system under item 1, wherein the storage device comprises a first storage subdevice intended for storing predetermined excitation signals, and the second storage subdevice to store additional excitation signals, and a speech encoder includes means for recording in the second storage subdevice outside breaks the weighted sum of the excitation signal lane is Uchenie pauses next signal, extracted from the speech signal.

3. A transmitter for transmitting speech signal containing a speech encoder having a storage device for storing excitation signals, the first filter associated with the storage device, for filtering the excitation signals, the selector associated with the first filter and the second filter, for comparing the signal derived from the speech signal through a second filter, with the output signal of the first filter and to select the optimum excitation signal, a detector for detecting pauses in speech and to control the switch that turns off at least part of the speech coder when it detects a pause in speech, characterized in that the switch is configured to turn off the picker and the storage device in case of delay, off of the first filter from the storage device and connected to the first filter further signal derived from the speech signal.

4. The transmitter under item 3, wherein the storage device comprises a first storage subdevice intended for storing predetermined excitation signals, and the second storage subdevice to hold up is the subdevice outside pauses weighted excitation signal of the first storage subdevice and balanced excitation signal of the second storage subdevice and writing in the second storage subdevice during the pauses of the next signal, extracted from the speech signal.

5. Mobile radio for voice signal transmission containing a transmitter that includes a speech encoder that contains a storage device for storing excitation signals, the first filter associated with the storage device, for filtering the excitation signals, the selector associated with the first filter and the second filter, for comparing the signal derived from the speech signal through a second filter, with the output signal of the first filter and to select the optimum excitation signal, a detector for detecting pauses in speech and to control the switch that turns off at least part of the speech decoder upon detection of a pause in speech, characterized in that that switch turns off the picker, and a storage device in the case of pauses in speech, disables the first filter from the storage device and connects the first filter to the next signal derived from the speech signal.

6. Mobile radio under item 5, wherein the storage device comprises a first storage subdevice intended for storing predetermined excitation signals, and the second storage podu the si in the second storage subdevice outside breaks the weighted sum of the excitation signal of the first storage subdevice and balanced excitation signal of the second storage subdevice and writing in the second storage subdevice during the pauses of the next signal, extracted from the speech signal.

7. The mode of transmission of speech signals, in which the stored signals of excitation in the storage device, the filtered signals of arousal, carry out the comparison signal derived from the speech signal with the filtered excitation signal and the choice of the optimal excitation signal, and perform detection of pauses in speech off at least part of the transmission system when it detects pauses, characterized in that when there is a pause in speech provide off picker and submission to the next filter signal derived from the speech signal.

8. The method according to p. 7, wherein the storage device comprises a first storage subdevice, which stores pre-defined excitation signals and the second storage subdevice, which store additional excitation signals, and the speech coder include recording media in the second storage subdevice outside breaks the weighted sum of the excitation signal of the first storage subdevice and balanced excitation signal of the second storage subdevice, and writing in the second storage subdevice during the

 

Same patents:

The invention relates to communication systems and particularly to systems in which communications are reused when terrestrial communications

The invention relates to the field of digital and pulse technique

The invention relates to the field of digital and pulse technique

The invention relates to the field of radio and, in particular, can be used for direction finding in mobile radio systems

The invention relates to a method for transmission of voice information, and the transmitter and receiver in accordance with the restrictive parts of paragraphs 1, 7 and 8 of the claims

The invention relates to a method for transmission of voice information, and the transmitter and receiver in accordance with the restrictive parts of paragraphs 1, 7 and 8 of the claims

The invention relates to communication systems and particularly to systems in which communications are reused when terrestrial communications

The invention relates to the field of digital and pulse technique

The invention relates to the field of digital and pulse technique

The invention relates to the field of radio and, in particular, can be used for direction finding in mobile radio systems

The invention relates to a method for transmission of voice information, and the transmitter and receiver in accordance with the restrictive parts of paragraphs 1, 7 and 8 of the claims

The invention relates to a method for transmission of voice information, and the transmitter and receiver in accordance with the restrictive parts of paragraphs 1, 7 and 8 of the claims

The invention relates to a method for transmission of voice information, and the transmitter and receiver in accordance with the restrictive parts of paragraphs 1, 7 and 8 of the claims
Up!