An audio signal recording and reproducing apparatus

The present invention relates to an audio signal recording and reproducing apparatus of the type for digitising sound into an electric audio signal, data compressing said signal so as to reduce information quantity, recording said compressed signal into a semi-conductor memory and reproducing an audio signal from said recorded data.

A known audio signal recording and reproducing apparatus, using differential PCM coding, is shown in Fig. 4. An input microphone 1 converts sound into an electrical audio signal. An input signal amplifier amplifies the audio signal from the input microphone and an input filter 3 removes unwanted frequency components and performs pre-emphasis. An analogue to digital converter 4 digitises the signal from the input filter 3. A new input data register 5 holds the digital data obtained by the analogue to digital converter 4 and an old input data register 6 holds the A-D converter digital data of the previous sample. An input data difference operator 7 operates a difference between the data held within the data registers 5 and 6. A memory interface 8 conducts processing, such as addressing, writing in, reading out and refresh operations, against a semi-conductor memory. A semi-conductor memory 9 receives the operation result obtained by the input data difference operator 7, which is written in through the memory interface 8.

An old output data register 11 holds the output data of one sample and an output data difference operator 10 reads out the difference data recorded in the semi-conductor memory 9, through the memory interface 8 and conducts an addition operation against the contact of the old output data register 11. A new output data register 12 holds the operation result of the output data difference operator 10 and a D-A converter 13 converts the digital data at the new output data register 12 to an analogue value. An output filter 14 removes quantization noise and executes emphasis thereto. An output signal power amplifier 15 amplifies the signal from the output filter 14 and a speaker 16 converts the electrical audio signal on the power amplifier 5-15 into sound.

The device operates as follows. At first, sound is converted into an electrical signal by the input microphone 1 and, thereafter, the audio signal is amplified by the input signal amplifier 2. Removal of extra-audio signal band frequency components and pre-emphasis are executed by the input filter 3. Thereafter, the signal is digitised by the A-D converter 4. The digitised audio signal is held by the new input data register 5 and the difference between that and the data of the old data input register 6 is operated by the input data difference operator 7 and the result is written into the semi-conductor memory 9 through the memory interface 8. The data of the new input data register 5 is transferred and stored by the old input data register 6, after the conclusion of operation by the input data difference operator 7.

The difference data stored in the semi-conductor memory 9 is read out by the memory interface 8 and difference operation is executed to the old output data register 11 by the output data difference operator 10. The operation result is stored in the new output data register 12 as well as being stored at the old output data register 11.

The output digital data stored in the new output data register 12 is converted into an analogue signal by the D-A converter 13 and removal of quantization noise and emphasis are executed thereto by the output filter 14. Thereafter, the result is power amplified by the output signal power amplifier 15 and is output as sound from the output speaker 16.

In the known recording and reproducing apparatus, using differential PCM, it is not possible to follow a change in audio signal which exceeds a predetermined range of difference values. This results in deterioration of sound quality of the reproduced sound, due to the compression of the audio information.

DE-A-3 534 064 discloses an audio signal recording and reproducing apparatus which digitises and compresses an input signal prior to recording it in a semiconductor memory, and reproduces an audio signal from the recorded data.

EP-A-0 117 525 describes an adaptive encoder which uses, for the purpose of data reduction, a plurality of conversion tables stored in a read-only memory and selected in accordance with the difference value of subsequent signal amplitudes.

It is an object of the present invention to provide an audio signal recording and reproducing apparatus capable of reproducing a higher quality of sound.

According to an aspect of the present invention, there is provided an audio signal recording and reproducing apparatus of the aforesaid type, characterised by a plurality of input compression and output extension conversion tables which are used for compression conversion of input signals and extension conversion of output signals respectively; and input compression and output extension conversion table selection counters count-controlled in response to difference data representing the change of the audio signal, which select one among a plurality of input compression and output extension conversion tables in accordance with their count value; wherein said tables represent an increasing range of difference values from a low range to a high range, and each table as an up-counting region indicating that a higher range of values is required for the next difference value and a down-counting region indicating that a lower range of values is required for the next difference value and the output counter is adjusted in response to the level read from the memory so that the output counter follows the input counter.

It will be appreciated that the invention as above defined can be applied to individual recording or reproducing apparatuses

The invention will now be described by way of example only, with reference to the accompanying drawings, in which:

• Fig. 1 is a diagram showing a construction of an audio signal recording and reproducing apparatus according to the present invention;
• Fig. 2 is a diagram showing constructions of the input compression conversion table selection counter and the input compression conversion tables for explaining a concrete operation of the input difference conversion section of the apparatus of Fig. 1.;
• Fig. 3 is a diagram showing a flow chart, for executing the operation of the embodiment of Fig. 1 under program control; and
• Fig. 4 is a diagram showing a construction of a known audio signal recording and reproducing apparatus.

An audio signal recording and reproducing apparatus is shown in Fig. 1, where referenced numeral 20 designates an input microphone, referenced numeral 21 designates an input signal amplifier, referenced numeral 22 designates an input filter, referenced numeral 23 designates an A-D converter, referenced numeral 24 designates a new input data register and reference numeral 25 designates an old input data register, and these have the same or corresponding functions as those already described in the prior art device. An input data difference operator 26 is one for operating a difference between the held data of the both registers 24 and 25. An input compression conversion table selection counter 27 is one for selecting one among a plurality of input compression conversion tables 29. An input difference comparison section 28 is one for comparing the operation result by the input difference operator 26 and the content of a table selected from a plurality of input compression conversion tables 29 by the input compression conversion table selection counter 27. A memory interface 30 is one for conducting writing in, reading out, and refresh operation against a semiconductor memory 31. An output extention conversion table selection counter 34 is one for selecting one among a plurality of output extention conversion tables 33. A register 32 is one for holding the output difference value obtained by the output extention conversion table which is selected by the output extention conversion table selection counter 34 and the memory interface 30. An output data difference operation section 35 is one for operating a difference between the contents of the register 32 and the old output data register 36. A new output data register 37 is one for holding the operation result of the output data difference operator 35. Reference numeral 38 designates a D-A converter, reference numeral 39 designates an output filter, reference numeral 40 designates an output signal power amplifier, and reference numeral 41 designates an output speaker, and these have the same or corresponding functions as those already described in the prior art device.

The device operates as follows.

First of all, sound is converted into an electric signal by the input microphone 20 , and thereafter the audio signal is amplified by the input signal amplifier 21. Removal of extra audio signal band frequency components and preemphasis are executed thereto by the input filter 22 and the result is converted into a digital value by the A-D converter 23. The audio data converted into digital data is held by the new input data register 24 and the difference between that and the content of the old input data register 25 is operated by the input data difference operator 26. The operation result is compression converted by the input difference conversion section 28 with the use of an input compression conversion table among the input compression conversion tables 29 which is selected by the input compression conversion table selection counter 27, and it is written in into the semiconductor memory 31 through the memory interface 30. The input compression conversion table selection counter 27 is up-counted or down-counted in accordance with the present time conversion value after the data to be written in into the semiconductor memory 31 is determined by the input difference conversion section 28, and thus a count value representing a new input compression conversion table is determined.

The data stored at the semiconductor memory 31 is read out by the memory interface 30, and it is converted into a corresponding difference value in a table among a plurality of output extention conversion tables 33 which is selected by the output extention conversion table selection counter 34, and thereafter a difference operation is executed to the content of the old output data register 36 by the output data difference operator 35, and the operation result is stored at the new output data register 37. The output extention conversion table selection counter 34 is up-counted or down-counted in accordance with the present time conversion value after the content of the output difference register 32 is determined and a count value representing a new extention conversion table is determined.

The data stored at the new output data register 37 is converted into an analogue signal by the D-A converter 38, and thereafter removal of quantization noise and emphasis are executed thereto by the output filter 39, and the result is power amplified by the output signal power amplifier 40 and is output as sound from the output speaker 41.

Next, an example of a method for converting the difference data between the contents of the new input data register 24 and the old input data register 25 into data to be stored in the semiconductor memory 31 will be described with reference to figure 2.

First of all, in the initial state, the input compression conversion table counter 50 (corresponding to 27 of figure 1) is in a reset state, and in this state, table No. 1 is selected by the output of the decoder 51. In this state, the differences from +4 to -3 correspond to the values of table No. 1 one to one and they are converted to conversion values of "7" to "0", respectively.

The difference data larger than +4 and the difference data smaller than -3 are converted into "7" and "0", respectively, as they are to be +4 and -3, respectively.

In this state, when a value larger than or equal to 3 and a value smaller than or equal to -3 are input as difference data, they are converted into corresponding conversion values as well as the input conversion table selection counter 50 is up-counted, thereby selecting table No. 2 as a conversion table for the next sample value.

The difference data of the m-th sample value is compression converted by utilizing table No. n which is determined by the change of the difference data from the initial state, and when the conversion data, that is, the conversion value is larger than or equal to 6 or smaller than or equal to 0, the input compression conversion table selection counter 50 is up-counted and when the conversion data is smaller than or equal to 5 or larger than or equal to 1, the input compression conversion table selection counter 50 is down-counted, thereby determining the compression conversion table for the next sample value.

When the conversion value is smaller than or equal to 5 or larger than or equal to 1 in a state where the input compression conversion table selection counter 50 represents table No. 1, the input compression conversion table selection counter 50 does not change. Similarly as above, when the conversion value is larger than or equal to 6 or smaller than or equal to 0 in a state where the input compression conversion table selection counter 50 represents table No. 6, the input compression conversion table selection counter 50 does not change.

The way of reading out data recorded in the semiconductor memory 31 and reproducing the new output data are conducted in accordance with the order reverse to that of the recording processing described above.

While in the above illustrated embodiment the recording section which digitizes an audio signal, data compresses the signal so as to reduce the information quantity, and writes in the signal into the semiconductor memory and the reproducing section which reads out information from the semiconductor memory and outputs sound are produced integrated with each other, the recording section and the reproducing section may be provided independently on each other.

Furthermore, the input data difference operator 26, the input compression conversion table selection counter 27, and the input compression conversion tables 29 may be those which also function as an output data difference operator 35, an output extention conversion table selection counter 34, and output extention conversion tables 33, respectively.

While figure 2 shows an example of conversion used in the present invention in which the conversion value has 3 bits and the table has 6 ways of conversion, the conversion value may have 4 bits and the table may have more than 6 ways of conversion.

While in the above illustrated embodiment the condition that the input compression conversion table selection counter is up-counted is set to that the conversion value is larger than or equal to 6 or smaller than or equal to 0, this condition may be any other combination of conditions that the conversion value is larger than or equal to 4 or smaller than or equal to 2. Similarly as above, while the condition that the input compression conversion table selection counter is down-counted is set to that the conversion value is smaller than or equal to 5 and larger than or equal to 1 in the above illustrated embodiment, this condition may be any combination of conditions that the conversion value is smaller than or equal to 6 and larger than or equal to 1.

The conversion values and the contents of the table may be any other values.

Furthermore, the operation of the above illustrated embodiment may be realized by a program corresponding to the flowchart shown in figure 3. In figure 3, reference characters S1 to S8 designate processing steps at the recording section and reference characters S11 to S18 designate processing steps at the reproducing section. In figure 3, the after-processing steps S6 and S16 are processing steps for renewing the address of the semiconductor memory after writing in or reading out operation and the conclusion checking steps S7 and S17 are steps for detecting that the address has reached the final address of the semiconductor memory or, in some cases, the final address of some block of the semiconductor memory, thereby to conclude the processing. The sampling synchronization steps S8 and S18 are steps for synchronization of sampling timing of the A/D or D/A conversion. Other than software processing for timing, a squeeze processing using a timer as a hardware processing can be employed.

As is evident from the foregoing description, according to the present invention, a plurality of input compression and output extention conversion tables are provided and the most appropriate compression and extention conversion tables are selected by the table selection counter in accordance with the change of the audio signal, whereby a high quality sound can be reproduced even from data of less information quantity. Furthermore, data conversion is conducted by a table, and when the device is realized by a hardware the replacement of table data is easened, thereby resulting in a low cost system. Even when it is realized by a software, the replacement of table data is easened, thereby shortening the processing time.