CN1173712A - System for recording and/or reproducing pulse code modulated digital audio signals - Google Patents

Abstract

The invention provides a digital audio recording apparatus for formatting digital audio data and recording formatted data on a digital recording medium, so that digital audio data having a first sampling frequency and other digital audio data having a second sampling frequency being twice the first sampling frequency can be reproduced in both a reproduction apparatus having an operating frequency for digital-to-analog converting the first sampling frequency and that having an operating frequency for digital-to-analog converting the second sampling frequency. The digital audio recording apparatus transforms the digital audio data into transformation coefficients in a frequency domain, the transformation coefficients are formatted using one of predetermined recording formats. Corresponding digital audio reproduction apparatus reproduces an analog audio signal from the formatted data, based on an operating frequency for D/A conversion, a formatting method of the formatted data, and a sampling frequency of the audio signal contained in the formatted data.

Description

Systems for recording and/or reproducing pulse code modulated digital audio signals

The present invention relates to a system for recording and/or reproducing pulse code modulated digital (PCM) audio signals, and more particularly to a system in which even though the sampling frequencies of the PCM digital audio signals are different, the digital audio signals can be A system for recording and/or reproducing PCM digital audio signals recorded on or reproducible from a digital recording medium.

All devices that use PCM digital audio signals use a reference unit, which is a data unit that enables error correction. The size of the reference unit is determined according to the characteristics of a transmission medium. A digital audio signal sampled according to a predetermined sampling frequency is recorded on a recording medium in the form of data blocks having the data recording characteristics of a recording medium such as a compact disc (CD), digital audio tape (DAT), etc. The determined data size or data volume. When a reproduction operation is performed with respect to the recording medium, the digital-to-analog converter converts a digital audio signal into an analog audio signal in the order of a data stream from the recording medium.

Regarding PCM digital audio signals, in order to replace the current system using a sampling frequency of 44.1KHz or 48KHz with a new audio recording system using a sampling frequency of 88.2KHz or 96KHz, for example, an advanced audio (ADA) system has been proposed in Japan . However, it is impossible to reproduce audio signals recorded in digital form according to the newly proposed system using existing reproduction systems having a sampling frequency of 44.1 KHz or 48 KHz. Therefore, in order for such an existing reproducing system to be able to reproduce digital audio signals recorded according to the new recording system, a separate digital signal processing is required so as to process the number of bits representing each sample without causing noise. In the case where digital signal processing is required, there is a low-pass filter to eliminate frequency components above 44.1KHz or 48KHz in the signal read from the recording medium, and there is a decimation filter to enable the low-pass The filtered signal enters a signal with a sampling frequency of 44.1KHz or 48KHz.

Accordingly, in order to reproduce an audio signal according to the newly proposed system, the existing reproduction system will process all input audio data. Therefore, there is a need for a digital processor whose operating frequency is greater than the sampling frequency of the input audio data. Therefore, the cost of modifying the existing reproducing system in order to reproduce the digital audio signal of the proposed signal recording system is too high. In addition, it is also necessary to increase the RAM capacity for the above-mentioned individual digital signal processing, so that an increase in cost is necessary for such an improvement.

Meanwhile, a reproduction system including a digital-to-analog converter with a sampling frequency of 88.2 KHz or 96 KHz directly transmits received data to an output without processing the received data. Therefore, a digital processor and a large-capacity RAM for processing audio data are not required. However, if another audio recording system using a frequency higher than 88.1KHz or (96KHz) is proposed, the same problems as the current system will arise.

In order to solve the above-mentioned problems, an object of the present invention is to provide a method for converting audio samples in a predetermined reference unit of a digital audio signal into transform coefficients in a frequency domain and formatting the transform coefficients according to a predetermined recording format. A system for recording audio signals sampled according to the current recording method and the new recording method in existing audio devices and new audio devices using the new recording method.

Another object of the present invention is to provide a method for converting audio samples in predetermined reference units of a digital audio signal into transform coefficients in a frequency domain and formatting the transform coefficients in accordance with a predetermined recording format in the current A system for reproducing audio signals sampled according to the current recording method and the new recording method in an audio device and a new audio device using a new recording method.

In order to achieve the above-mentioned purpose of the present invention, a kind of digital audio recording device for a digital recording medium is provided, and this digital audio recording device comprises:

A reference unit formatter for receiving a pulse code modulated digital audio signal and forming a sequence of reference units, each of which has a predetermined size from the received digital audio signal;

Transformation means receiving the sequence of reference units for transforming audio samples belonging to an individual reference unit into transform coefficients in a frequency domain and outputting a sequence of reference units comprising transform coefficients;

Formatting means for receiving the sequence of reference units from the transformation means, and using one of the first record format and the second record format to format and output the received sequence of reference units contained in the respective Among the transform coefficients in the reference units, the first recording format is used to form a first recording unit including transform coefficients belonging to the respective reference units, and the second recording format is used to form a second recording unit including transform coefficients belonging to a low frequency region and a third recording unit including transform coefficients belonging to a high frequency region; and

The recording device is used for recording the recording unit sequence output from the formatting device on the digital recording medium.

In order to achieve another purpose of the present invention, a digital audio reproduction device provided includes:

an input terminal for receiving data formatted in one of a first recording format and a second recording format, wherein among the transform coefficients contained in each reference unit, the first recording format is used for forming a first record unit including transform coefficients belonging to each reference unit, and a second record format for forming a second record unit of transform coefficients included in a low frequency region and transform coefficients included in a high frequency region The third recording unit of ;

A digital-to-analog (D/A) converter for digital-to-analog conversion of an input digital signal according to a predetermined operating frequency;

means connected to the input terminal for identifying the sampling frequency and a recording format of the audio signal contained in the input formatted data, and outputting data representing the identification result;

Deformatting means, connected to the input and receiving data output from the means for identification, for deformatting the received data based on a comparison between the identified sampling frequency and a predetermined operating frequency deformatting the formatted data, and deformatting the identified recording formats of the pre-stored first and second recording formats; and

inverse transform means having an inverse transform characteristic optimized for the D/A converter for transforming the transform coefficients output from the inverse format means into audio samples in a time domain, and supplying the D/A converter Provides a sample of the generated audio.

A preferred embodiment will be described below with reference to the accompanying drawings.

Fig. 1 is a schematic block diagram of a digital audio recording device according to a preferred embodiment of the present invention;

Figure 2 is a detailed block diagram of the formatting unit shown in Figure 1;

3A to 3D are used to explain the recording format according to the present invention;

Fig. 4 is a schematic block diagram of a digital audio reproduction device according to the present invention; and

Fig. 5 is a block diagram of a digital audio reproducing apparatus according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

The digital audio signal recording device according to the present invention shown in Fig. 1 can record the audio signal with 44.1KHz or 48KHz sampling frequency according to the recording format proposed in the present invention on a recording medium according to the existing audio recording method and according to A new audio recording method records audio signals with a sampling frequency of 88.2KHz or 96KHz. For the sake of clarity, the relatively higher sampling frequency is referred to as the first sampling frequency, and the relatively lower sampling frequency is referred to as the second sampling frequency.

An audio signal sampled in the time domain quantized using linear PCM is input into the reference unit former 10 of FIG. 1 . The reference unit former 10 divides the input digital audio signal into a plurality of reference units each having a predetermined size. The described reference units and recording units are determined by an audio quality and the characteristics of the transmission medium. In other words, the size of the reference cell is determined according to the amount of data by which errors imposed during recording and reproduction of the linear pulse code modulated digital audio signal can be corrected. The size of the recording unit is determined based on a data amount that can be continuously recorded on or read from a digital recording medium according to signal recording characteristics of the digital recording medium. Figure 3A shows an example of a reference cell obtained by separation. The reference unit of Figure 3A consists of a sequence of 2m audio samples. The reference cell former 10 provides reference cells obtained by sequentially partitioning the transformers 20 . A transformer 20 transforms the audio samples into transform coefficients in a frequency domain. The performance of the converter 20 is determined by the audio quality of a target audio signal. The transform coefficients obtained by the domain transform of the respective reference units are output to the formatting unit 30 in a form that can identify each reference unit. The formatting unit 30 formats the transform coefficients in each reference unit by using one of the first recording format shown in FIG. 3B and the second recording format shown in FIGS. 3C and 3D . The detailed structure and operation of the formatting unit 30 will be described in detail with reference to FIG. 2 and FIGS. 3B to 3D. The recorder 40 digitally records the data formatted in the formatting unit 30 on a digital recording medium (not shown) such as CD and DAT.

Referring to FIG. 2 , FIG. 2 shows a detailed block diagram of the formatting unit 30 . The recording format indicator 32 in the formatting unit 30 designates one of the first and second recording formats corresponding to the reference unit obtained in the reference unit former 10 . The recording format indication of the recording format indicator 32 can be done by design by a user who wants to record audio signals.

The recording format indicator 32 outputs to the first data formatter 34 a sequence of transform coefficients received from the transformer 20 when the first recording format is selected. The first data formatter 34 formats the audio samples within a single reference unit into first record units and outputs the first sequence of record units to the first additional data inserter 35 . The first recording unit includes transform coefficients obtained from a reference unit.

FIG. 3B shows transform coefficients F[0], F[1], F[2], . . . , and F[2m] contained in the first recording unit. In FIG. 3B, F[0] represents the lowest frequency transform coefficient, and F[2m] represents the highest frequency transform coefficient. The first additional data inserter 35 inserts, into The start of this record unit sequence. Because the present invention is not limited to the example of inserting the first and second additional data into the beginning of the recording unit sequence, it is clear to those skilled in the art that the first data formatter 34 can be implemented in such a way Modification, that is, the first and second additional data may be appropriately inserted in the sequence of recording units according to the characteristics of a transmission channel or a recording medium.

When the second recording format is selected, then the recording format indicator 32 outputs the sequence of reference units received from the converter 20 to the second data formatter 36 . The second data formatter 36 formats the transform coefficients within a single reference unit provided from the record format indicator 32 into two record units. In detail, the second data formatter 36 equally divides the transform coefficients in the reference unit, and forms a second recording unit composed of the transform coefficients in the relatively lower frequency region and in the relatively higher frequency region The transformation coefficients constitute the third recording unit. Therefore, two recording units are obtained from a reference unit. FIG. 3C shows transform coefficients F[0], F[1], F[2], . . . , and F[m] contained in the second recording unit. FIG. 3D shows transform coefficients F[m], F[m+1], F[m+2], . . . , and F[2m] contained in the third recording unit. The second data formatter 36 outputs to the second additional data inserter 37 a sequence of recording units derived from the sequence of reference units. In the recording unit sequence, the second and third recording units corresponding to a relatively time-advanced reference unit are temporally ahead of the second and third recording units corresponding to a relatively time-lag reference unit. In the case of recording units corresponding to the same reference unit, the second recording unit is temporally ahead of the third recording unit.

The second additional data inserter 37 receives a sequence of the recording unit from the second data formatter 36 and inserts the aforementioned first additional data and second additional data into the output from the second data formatter 36. The beginning of a sequence of record units. In addition, the second additional data inserter 37 inserts third additional data for identifying the second and third recording units obtained from the same reference unit into respective start portions of the second and third recording units.

The above-described arrangement of Fig. 1 can use two recording formats. However, in practical embodiments it is preferable to use only one recording format. If the apparatus of FIG. 1 uses two recording formats, separate additional data for identifying the two recording formats, ie, recording format data, is required. In this case, the recorder 40 in the apparatus of FIG. 1 inserts the recording format data at the beginning of the formatted data.

The reproducing apparatus of FIG. 4 reproduces a digital audio signal from a recording medium (not shown) on which a digital audio signal is recorded, and the apparatus includes a D/A converter 81 having the same operating frequency as the second sampling frequency. . The reader 50 reads formatted data from the recording medium (not shown) on which the digital audio signal converted into a frequency domain is recorded. The audio signal included in the formatted data may have a first sampling frequency or a second sampling frequency as described above. In the reader 50, the frequency identifier 51 identifies the sampling frequency of a digital audio signal based on the first additional data included in the read formatted data, and outputs the identified frequency to the recording format identifier 53. result. The recording format recognizer 53 uses the additional data contained in the formatted data received from the frequency recognizer 51 to recognize the recording format of the formatted data. If the signal recorded in the recording medium is data formatted according to the first recording format, the data includes first and second additional data and recording format data. If the signal recorded on the recording medium is data formatted according to a second recording format, the data includes first, second and third additional data and recording format data. The reader 50 supplies the second and third additional data and recording unit sequences based on the frequency identification result and the identified recording format to the deformatting unit 60(1). In order to provide data, it can be designed that the reader 50 provides all the data read from the recording medium to the unformat unit 60(1), and the unformat unit 60(1) based on the data received from the reader 50 data to identify the sampling frequency and recording format and selectively supply audio data to the inverse converter 71 .

The deformatting unit 60 ( 1 ) provides the data received from the reader 50 to the first data extractor 61 when the data recorded on the recording medium has a first sampling frequency and a first recording format. For each first recording unit of data received from the reader 50, the first data extractor 61 extracts transform coefficients in the low frequency region. In more detail, the first data extractor 61 extracts the transform coefficients of the low-frequency region from the transform coefficients in the separate recording unit based on the second additional data included in the data received from the reader 50 and extracts the transform coefficients of the low frequency region from the received data Delete the second additional data in . Here, the low frequency region means a lower frequency region of two frequency regions equally divided from the entire frequency region represented by all transform coefficients in a single recording unit. As a result, the individual first recording unit data output from the reader 50 is decimated by 1-in-2 (2:1) by means of the first data extractor 61 . The deformatting unit 60 ( 1 ) provides the data received from the reader 50 to the first data extractor 61 when the data recorded on the recording medium has the first sampling frequency and the second recording format. The first data extractor 61 extracts the second recording unit based on the third additional data contained in the data received from the reader 50, and deletes the third additional data. When the data recorded on the recording medium has the second sampling frequency and the first recording format, the deformatting unit 60(1) deletes the second additional data contained in the data received from the reader 50. When the data recorded on the recording medium has a second sampling frequency and a second recording format, the first deformatter unit 60(1) provides a first deformatter 63 with the data from the reader 50 received data. The first de-formatter 63 de-formats the second and third recording units corresponding to the same reference unit based on the pre-stored second recording format and third additional data. Then, the first inverse formatter 63 deletes the third additional data and outputs the finally remaining transform to the inverse transformer 71 according to the above-mentioned signal processing.

Because the present invention is not limited to the situation in which the first data extractor 61 and the first deformatter 63 mentioned in this example delete the second and/or third additional data, it should be clear to those skilled in the art that The unformatting unit 60(1) is modified to delete the second and third additional data.

In addition, those of ordinary skill in the art should also be clear about this modification, that is, the reader 50 outputs all the data read from the recording medium to the reverse format unit 60(1), and the reverse format unit 60(1) The record format data in the data received from the reader 50 is deleted.

The inverse transformer 71 is suitably designed to inversely transform a single reference unit transform coefficient obtained from audio samples having a second sampling frequency, and the inverse transformer 71 will receive from the deformatting unit 60(1) The transform coefficients are inverse transformed into audio samples in a time domain. Audio samples output from the inverse converter 71 are digital-to-analog converted in the D/A converter 81 . 4 can reproduce an audio signal having a sampling frequency corresponding to twice an operating frequency of the D/A converter 81 and an audio signal having a sampling frequency corresponding to an operating frequency of the D/A converter 81. audio signal. Therefore, when the audio signal recorded on the recording medium has the first sampling frequency and the D/A converter 81 operates according to the second sampling frequency, the device of FIG. Sound quality audio signal.

Unlike the arrangement of Fig. 4, Fig. 5 shows an arrangement comprising a D/A converter 83 having the same operating frequency as the first sampling frequency. In FIG. 5, since blocks having the same reference numerals as in FIG. 4 perform the same functions as the blocks in FIG. 4, detailed descriptions are omitted. The unformat unit 60(2) uses the data from the reader 50 to perform an unformat operation based on the formatted data. The deformatting unit 60 ( 2 ) provides the data received from the reader 50 to the second data extractor 65 when the data recorded on the recording medium has the first sampling frequency and the first recording format. The second data extractor 65 extracts the first record unit contained in the data received from the reader 50 and then deletes the second additional data. Thereafter, the inverse format unit 60( 2 ) outputs the frequency identification result and the first recording unit to the inverse transform unit 73 . The inverse converter 73 determines that the first record unit received at present has the same sampling frequency as a sampling frequency in the D/A converter 83 according to the frequency identification result, and the conversion coefficient contained in the first record unit alone will be Transform to audio samples in a time domain. When the data recorded on the recording medium has the second sampling frequency and the first recording format, the second data extractor 65 in the deformatting unit 60(2) deletes the data contained in the slave reader 50. The second additional data among the received data outputs the first record unit and the frequency identification result to the inverse converter 73 . In this case, if the transform coefficients belonging to the first recording unit are transformed into audio samples of the time domain, the number of audio samples becomes smaller than the number of audio samples belonging to the reference unit of signal processing related to the apparatus of FIG. 5 . For this reason, the data padding unit 75 in the inverse transformer 73 performs zero padding to pad the data in the frequency domain. The inverse transformer 73 transforms the zero-padded transform coefficients of the first recording unit into audio samples in the time domain.

When the data recorded on a recording medium has a first sampling frequency and a second recording format, the second de-formatter 67 in the de-formatting unit 60(2) according to a pre-stored second The record format deformats the data received from the reader 50 . In more detail, the second deformatter 67 deformats the second and third recording units corresponding to the same reference unit into a corresponding reference unit according to the third additional data and the second recording format. Afterwards, the inverse format unit 60(2) deletes the third additional data and the recording format data, and outputs the reference cell sequence and the frequency identification result to the inverse converter 73. The inverse transformer 73 transforms the transform coefficients of each reference unit into audio samples in the time domain according to the frequency identification result received from the inverse formatting unit 60(2).

When the data recorded on the recording medium has a second sampling frequency and a second recording format, the second deformatter 67 performs a pairing of the data obtained from the reader 50 according to the third additional data and the pre-stored second recording format. The received second and third recording units are deformatted. In this case, the number of audio samples in the time domain becomes smaller than the number of audio samples required for signal processing by the apparatus of FIG. 5 . The data padding unit 135 performs the above-mentioned zero padding, and its resulting reference units are transformed into audio samples in the time domain in the inverse transformer 133 . These audio samples are digital-to-analog converted in the D/A converter 143 . Therefore, the apparatus of FIG. 5 can provide a digital audio signal having the current sampling frequency with an auditory signal to the listener having the same sound quality as a signal of the sampling frequency according to the new recording method.

The inverse transformer 73 of FIG. 5 may be designed to inverse transform the transform coefficients of the respective reference units obtained from the audio samples having a second sampling frequency. In this case, the inverse converter 73 can judge whether the data padding is performed without the frequency identification result of the inverse formatting unit 60(2).

Assuming that each reference unit includes K transform coefficients, the first sampling frequency is L and the second sampling frequency is M, then it can be calculated according to the formula Int(K×M)/L with an operation corresponding to the second sampling frequency The number of transform coefficients required in the current reproduction device of the frequency. Here, "Int" is an operator for discarding a value smaller than an integer value. In this case, transform coefficients having numbers obtained by the above formula are transformed into audio samples of the time domain. Using this principle, the system according to the present invention can be applied to an audio device in which the sampling frequency of an audio signal is increased to an integer multiple of the first sampling frequency.

The above embodiment has been described in conjunction with the first and second sampling frequencies. However, the present invention is also applicable to an audio signal having a sampling frequency greater than the first sampling frequency.

As described above, the audio signal recording and reproducing system according to the present invention can reproduce an audio signal having a frequency difference of twice between an audio device having a relatively low sampling frequency and an audio device having a relatively high sampling frequency.

In addition, audio signals having different sampling frequencies may be reproduced in an audio device having a relatively low sampling frequency and an audio device having a relatively high sampling frequency. In addition, the present invention can be further applied to an audio device employing a new sampling frequency increased by an integer multiple.

Claims (15)

1. A digital audio recording device for a digital recording medium, the digital audio recording device comprising: A reference unit forming device, used to receive a pulse code modulated digital audio signal, and form a sequence of reference units from the received digital audio signal, each reference unit having a predetermined size; Transformation means receiving the sequence of reference units for transforming the audio samples belonging to the respective reference unit into transform coefficients in a frequency domain and outputting a sequence of reference units comprising the transform coefficients; Formatting means for receiving the sequence of reference units from said transforming means, and using one of the first recording format and the second recording format to format and output the received reference unit sequence, wherein contained in the respective Among the transform coefficients in the reference unit, the first recording format is used to form a first recording unit including transform coefficients belonging to the respective reference units, and the second recording format is used to form a first recording unit including transform coefficients belonging to a lower frequency region. two recording units and a third recording unit containing transform coefficients belonging to the higher frequency region; and recording means for recording the sequence of recording units output from said formatting means on the digital recording medium. 2. The digital audio recording apparatus according to claim 1, wherein said lower frequency region and said higher frequency region are equally divided according to a frequency band represented by a transform coefficient. 3. The digital audio recording apparatus according to claim 1, wherein the predetermined size is determined according to an amount of data by which an error imposed during recording and reproducing a linear pulse code modulated digital audio signal can be corrected. 4. The digital audio recording apparatus according to claim 1, wherein the recording unit is determined based on an amount of data that can be continuously recorded on a digital recording medium according to a signal recording characteristic of the digital recording medium. 5. The digital audio recording apparatus as claimed in claim 1, wherein said formatting means further comprises an additional data inserter for representing a sequence of first recording units generated using the first recording format to be included in Additional data of the sampling frequency of the audio signal in the sequence of recording units is inserted into the sequence of recording units. 6. The digital audio recording apparatus according to claim 5, wherein said additional data inserter further inserts additional data indicating the number of audio samples in the respective recording units into the sequence of recording units. 7. The digital audio recording apparatus as claimed in claim 1, wherein said formatting means includes an additional data inserter for inserting a first sampling frequency representing a sampling frequency of audio signals contained in a recording unit sequence. additional data is inserted into the sequence of recording units, and for the sequence of recording units generated using the second recording format, inserting a second additional data identifying the first recording unit and the second recording unit derived from the same reference unit to the beginning of each first and second recording unit. 8. The digital audio recording apparatus according to claim 1, wherein said formatting means further inserts recording format data for identifying the first recording format and the second recording format into said recording unit sequence. 9. A digital audio reproduction device comprising: an input terminal for receiving data formatted using one of a first recording format and a second recording format for forming a data format containing transform coefficients belonging to each reference unit The first recording unit, the second recording format are a second recording unit for forming transform coefficients contained in a lower frequency region among transform coefficients contained in each reference unit and contained in a higher frequency band region A third recording unit of the transformation coefficient; a digital-to-analog (D/A) converter for digital-to-analog conversion of an input digital audio signal according to a predetermined operating frequency; means connected to said input terminal for identifying a sampling frequency and a recording format of an audio signal contained in the input formatted data and outputting data representing the identification result; de-formatting means connected to said input and receiving data from the output of the means for identifying for performing formatted data received on the basis of a comparison between the identified sampling frequency and a predetermined operating frequency de-formatting, and de-formatting the identified recording formats of the pre-stored first and second recording formats; and inverse transform means having an inverse transform characteristic optimal for said D/A converter for transforming the transform coefficients output from said inverse format means into audio samples in a time domain, and to said D A/A converters provide the resulting audio samples. 10. The digital audio reproduction apparatus of claim 9, wherein said input receives said formatted data from the digital recording medium. 11. The digital audio reproducing apparatus as claimed in claim 9, wherein said each recording unit is determined based on an amount of data that can be continuously recorded on said digital recording medium according to a signal recording characteristic of the digital recording medium . 12. The digital audio reproducing apparatus according to claim 9, wherein the lower frequency region and the higher frequency region are equally divided from the entire frequency region represented by the transform coefficient. 13. The digital audio reproducing apparatus as claimed in claim 9, wherein in the case where the identified sampling frequency is the predetermined operating frequency multiplied by a factor of 2, said inverse formatting means provides said inverse transform means with Transform coefficients of the low frequency region among the transform coefficients of the first recording unit, and the identified recording format is the first recording format. 14. The digital audio reproducing apparatus as claimed in claim 9, wherein in the case where the identified sampling frequency is the predetermined operating frequency multiplied by a factor of 2, said inverse formatting means provides said inverse transform means with Transform coefficients in a second recording unit, and the identified recording format is the second recording format. 15. A digital audio reproduction apparatus as claimed in claim 9, wherein said inverse formatting means provides all transform coefficients to said inverse transform means when said predetermined operating frequency is the identified frequency multiplied by a factor of 2, and The inverse transform means performs zero padding for each reference unit before inverse transforming the transform coefficients received from the inverse transform means.

Images