JPS5917912B2 - automatic broadcasting device - Google Patents

Abstract

PURPOSE:To secure the output of the signals corresponding to the prescribed broadcast sentence by dividing the intermediate recording part into the group into which the aural signals sent from the aural recording part are written with an optional ratio and the group from which the aural signals are read out for control and then reading the aural signals in sequence from the intermediate recording part. CONSTITUTION:Aural recording part 100 consists of the memory which records the aural signals of the contents required for broadcasting by dividing those signals into the fixed syllables, words and the like and also can give the independent writing and reading control to at least more than three units of the aural signals, and possesses the intermediate memory groups featuring the general memory capacity shown in the equation. On the other hand, intermediate recording part 200 gives the control to the aural signals sent from part 100 by dividing them into the group into which they are written with an optional ratio and the group from which they are read out. Thus the aural signals are read out sequentially from part 200 to deliver 300 the signals corresponding to the prescribed broadcast sentence.

Description

[Detailed Description of the Invention] This invention sequentially reads out necessary audio signals from a large number of digital audio signals recorded in a memory, and stitches these signals together to edit and broadcast a series of broadcast sentences. This invention relates to an automatic broadcasting device.

Conventionally, in this type of automatic broadcasting apparatus, there has been one shown in FIG. 1 as a parifer memory type of the reproducing section.

That is, in FIG. 1, 100 is an audio recording unit that records digital audio signals necessary for editing (hereinafter simply referred to as audio signals), 200 is an intermediary recording unit (buffer memory), and a pair of memories 210, 211 and these memories 210,
an input switch 201 for controlling writing of necessary audio signals from the audio recording unit 100 to the memory 210;
It has an output switch 203 that controls reading from 211.

300 is an output unit that outputs the audio signals read from the memories 210 and 211 as audio.

When using this device to sequentially read out the necessary audio signals from a large number of recorded audio signals and edit them into a single broadcast sentence, it is necessary to continuously reproduce the audio signals to be read out without any breaks. The minimum memory capacity WT of the buffer memory 200 can be expressed by the following equation (1).

WT=2・・TaCbiO...(1)1−(Vo
/Vi): One.

That is, if the memory capacities of the memories 210 and 211 are respectively WCbit], then the memory read time W/
0Csec], accesses the audio recording unit 100 (maximum access time TaCsec) and calls the audio signal,
It is necessary to write it into the memory 210 (or 211) (writing time W/Vi [SecU), and the following must be satisfied.

According to this equation (2), the minimum required memory capacity of each of the memories 210 and 211 is ↓ \▼υ/ ▼↓ノ.

In this case, since the memories 210 and 211 are accessed alternately, by doubling this equation (3), the total memory capacity WT (=2W) in equation (4) is obtained. In the case of the device shown in Figure 1,
The memory capacities of the memories 210 and 211 are both 1 [Bit]
, R1'B,-, TaCbit].

In such a conventional device, if the broadcast text to be edited is "This train is stopping ~" as shown in Figure 2a, the broadcast text in this case is the audio signal Sa, Sb, Sc trigram and It will be formed of Sd.

Note that b and c in FIG. 2 refer to the memories 210 and 21.
FIG. 1 is a diagram illustrating timings of writing and reading operations of 1; First person switch 210 and output switch 2
By connecting 03 as shown in FIG. empty signal) is read out (before TO).

Next, at the time of TO, input switch 201 and output switch 2
03, write the audio signal Sa2 to the memory 211, and write the audio signal Sa2 written to the memory 210.
Read l. Then, at the next time point Ti, the input switch 201 and the output switch 203 are turned on again, the audio signal Sa3 is written into the memory 210, and the audio signal Sa2 is read from the memory 211. By continuously performing the above operations, the audio signals Sa-Sd are edited in sequence, and broadcast texts of a required length are continuously read out and reproduced. However, in the case of such a conventional device, the memory 21
0,211 memory capacity W, its control unit (the minimum memory unit when reading or writing preferences, and corresponds to the memory W1 capacity W (11) of memories 210, 211), and audio recording Depending on the relationship with the signal length of the audio signal recorded in the section 100, unnecessary breaks (defective pauses) may occur, which is not preferable.

That is, as shown in FIG. 2, the audio signals Sa, Sb, Sd (
7) If the signal length is shorter than the control single range WTW (=-=) [Bit] as in the Sa3, Sb2, and Sd2 parts,
These audio signals are only written to two parts of the memories 210 and 211, and the remaining part of the memory has an empty part 210.
x, 211x will be formed.

Since the next audio signal will be newly written to the memory on the other side, the blank areas 210x and 21 of these memories will be
1x becomes a defective pose, will it be regenerated? This significantly deteriorates the naturalness of the voice produced. As a means to solve this problem, a method (fixed length memory method) in which the signal length of the audio signal is an integral multiple of [Bit] can be considered, but when recording a large number of audio signals in the audio recording section 100, must be adjusted to a predetermined signal length, which requires a great deal of effort. The purpose of this invention is to eliminate the above-mentioned drawbacks of conventional devices, and to provide an automatic broadcasting device that does not cause defective pauses even when the audio signal to be edited has an arbitrary length. It is something to do.

An embodiment of the present invention will be described in detail below with reference to the drawings. That is, in FIG. 3, 100 is an audio recording section, and 300 is an output section, which are the same as the conventional apparatus shown in FIG.

Reference numeral 200 denotes an intermediary recording unit having an intermediary memory group consisting of small-capacity units 220a to 220n, input switches 201a to 201n1, and output switches 203a to 203n.

Here, the intermediate memory group consisting of the memories 220a to 220n is divided into two memory groups 220α and 220β (in FIG. 3, for convenience, the memories 220a to 220g and the memory 220h
220n) and each memory group 220α,
Let the memory capacity of the memory 220β be the capacity of the memory 220α minus the capacity of the memory 220β l\v▲vノ ~q.

Therefore, the sum of the capacities of the memory group 220α and the memory group 220β is 2VI-NTCbiO, which is the total capacity of the intermediate memory group. Here, each memory group 220α, 22
Considering the relationship between the memory capacity φ of 0β and its write/read operations, the memory capacity for the write is the first
If it is possible to double the conventional one shown in the figure, then the minimum necessary memory capacity of each memory group 220α, 220β is as follows.

Equation (5) above is the time v! to read WCbit] from the intermediate memory group. /VO indicates that the next audio signal can be accessed and 2WCbits of that audio signal can be written into the intermediate memory group.

Therefore, for example, the signal length of the audio signal is Wx (W<Wx
○ In the case of 2W), it is possible to write Wx in one access and read W out of the written Wx, so the remaining memory area (2W - Wx) and the memory blank after reading The next audio signal can be written following the section.

In this case, the intermediate memory group in FIG. 4, that is, the memories 220a to 220 of each memory group 220α, 220β
It is assumed that the bit length of n is set (subdivided) to the control unit length of memory control. Further, the minimum signal length of the audio signal recorded in the audio recording unit 100 is
It is assumed that the bit length of the memories 220a to 220n is subdivided (for example, in units of bits) so that the memory capacity W of the memory 220β is larger than the memory capacity W of the memory 220β, and the signal length of each audio signal is an integral multiple of the control unit length of the memory. Further, write/read control of these intermediate memory groups 220a to 220n is performed by
The input switches 201a to 201n and the output switch 2 are configured to distinguish between writing and reading in an arbitrary ratio according to the length of each reproduced audio signal.
03a to 203n shall be controlled. Fig. 4 is an explanatory diagram for explaining the operation of Fig. 3, and Fig. 4 a shows the content of the broadcast text to be broadcast by editing each audio signal. It is the same as the conventional one.
Furthermore, FIGS. 4b and 4c show each memory group 220α, 220
It represents the timing of the write operation and read operation of β. In Figure 3, now the memory group 220
It is assumed that the audio signal Sa (DSa,
previous state).

In this state, when the TO point is reached, the memory group 220
The audio signal Sal written to α is output to the output switch 203.
a to 203g, and the next Sa2 is read to the memory group 22 via human switches 203h to 203n.
Written to 0β.

Next, the reading of the memory group 220α is completed. 'That is, when the time point Ti is reached, the audio signal Sa2 written in the memory group 220β at the TO timing will be read out via the output switches 203h to 203n, and the S
a3 is written to the memory group 220α via the input switches 201a to 201g.

In this case, the signal length of Sa3 is also smaller than the bit length W of the memory group 220α, so at this point the signal length of the memory group 220α
It cannot be filled in and has a blank space. Then,
When the reading of Sa2 written in the memory group 220β is completed, that is, when the time T2' is reached, Sa3 written in the memory group 220α is sequentially read out and the previous t1 is read out.
The first half of the next audio signal Sb(7)Sbl is sequentially written into the blank space of the memory group 220α that was not filled at the time '.

When this memory group 220α becomes full, the latter half (remaining part) of Sb and Sb2 are subsequently transferred to the memory group 220α.
Write sequentially to 20β. In this case as well, memory group 2
20β is not full and has a blank portion, but this blank portion will be filled at the next timing T3'. Note that the audio signals read out from the memory group 220α during this period T2Lt3' are the first half of Sa3 and Sbl.

Here, the write speed Vi to the memory groups 220α and 220β is considerably faster than the read speed VO (for example, Vi-2M ('Bit/Sec), VO=64KC
bit/Sec], ), and the next audio signal (in Fig. 4, audio signals Sa, Sb, Sc, (corresponding to Sd) (maximum Gd access time Ta) to prepare for writing at the next timing.

Memory group T during the timing period t1' and T2' described above.
This is the reason why the next audio signals Sdl and SCl cannot be written in the same timing period (successively) despite the presence of blank spaces in 2', T3' 220α and 220β. This is because access time is required for calling, and in such a case, the writing that would otherwise occur will be performed at the next timing.

It should be noted that if one audio signal, for example Sa, is formed by
, sa2, and Sa3 are not subject to such restrictions, and continuous writing is possible. The memory group 220α is stored in the same manner for T3' and subsequent ones.
, 220β are subjected to writing control so as to sequentially fill in the blank portions, and the read audio signal has no blank time or defective pause, and natural audio is reproduced.

By the way, in the case of the conventional device shown in FIG. 1, for example, the memory 2
Even if the blank parts 210x and 211x of 10 and 211 are created, the next audio signal is newly stored in the memory 210 or 2 on the other side.
11, and 210x and 211x remain blank. However, according to the device of the present invention, the memory group 220α and the memory group 220β shown in FIG. For example, the memory groups 220α and 22
Even if writing of the audio signal is completed in a part (front part) of 0β and a blank part remains in the remaining part (rear part), there is time to access the next audio signal and write that signal, so the remaining part of the writing will be completed. It does not remain blank like in conventional devices.

The above operation assumes that all audio signal accesses require the maximum access time Ta ('Sec), but if the audio signal access is performed in a shorter time than Ta (Ta' [Sec]), the following The behavior is as follows.

That is, if the access is possible after the maximum access time TaCsec] or the short time TlCsec, the memory 22
The writable area in 0α and 220β is Ta′・0
There is only Cbi0, and the next write is TaCsec
]It is necessary to wait, but writable Tl/0Cbit
] At the time of writing, the memory 220α or 220β
Excluding the second part of the memory that is being read inside, the remaining part is approximately 2
Ta. An audio signal is written in V0 [Bit] (most of the memories 220α and 220β).

Therefore, after Ta'[Sec], Ta('Sec]
Even if the waiting time occurs, there is still no problem as the audio signal of about Ta-VOCbit has already been recorded and remains. That is, if there is a normal audio signal to be written (when accessed,
Empty memory section (if the audio signal is too long) Memory 22
The blank area between 0α and 220β is the maximum Ta. It is VO [Bit]. Conversely, there is always more than about Ta-VOCbit of audio signals that can be read out from the memories 220α and 220β. Further, the audio signal may be, for example, the audio signal Sa shown in FIG.
3, W3=W2/2 [Bit] For smaller signals, the audio signal is short even though there is time to write it, so the audio signal that can be read is Ta. OCbi
t] There is nothing left, and in the worst case, it will be equivalent to the audio signal Sa3. However, since this equivalent amount is a margin after securing the access time and write time for the next audio signal Sbl to be written, at time T2' when reading out this audio signal Sa3, the blank area of the memory group 220α is and the memory group 220β following it stores the next audio signal S.
bl is written and no hexagram b defective pause occurs. As described above, according to the present invention, there is a group consisting of a plurality of unit memories each having at least a memory capacity and capable of independently controlling writing and reading, and into which audio signals from an audio recording section are written. Since it is equipped with an intermediary recording section that has an intermediary memory group that is controlled separately for the written audio signal and the readout group, there is no longer a long pause in the middle of a sentence as in conventional devices. It can reproduce extremely natural sounds.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a conventional device, FIG. 2 is a diagram explaining the operation of the conventional device, FIG. 3 is a block diagram of a device according to an embodiment of the present invention, and FIG. 4 is a diagram explaining the operation of the device according to the present invention. It is a diagram. In the figure, 100 is an audio recording section, 200 is an intermediary recording section,
201a to 201n are input switches, 220a to 220
n is a unit memory, 203a to 203n are output switches,
300 is an output section.

Claims (1)

[Scope of Claims] 1. An audio recording system that records an audio signal of content necessary for broadcasting by dividing it into predetermined syllables or words, etc., which is formed by a plurality of unit memories that can independently control writing and reading, and whose total Memory capacity is at least 2[Vo]/[1-2(Vo
/Vi)]・Ta [bit] However, [Vo; Speed of reading from memory [bit/sec] Vi; Speed of writing to memory [bit/sec] ¥Vo<Vi¥ Ta; Maximum access time [sec]] an intermediate memory group, and each unit memory of the intermediate memory group is divided into a group into which data is written according to the signal length of the audio signal accessed from the audio recording section and a group from which the written audio signal is read out. An intermediary recording section that is controlled separately;
An automatic broadcasting device comprising an output section that sequentially reads audio signals from the intermediary recording section and reproduces and outputs them as a series of broadcast sentences.