JP3267193B2 - Voice reading device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technology for reading aloud by voice by specifying the number of words per unit time.

[0002]

2. Description of the Related Art In audio or video equipment such as a tape recorder, when fast forward is selected during reproduction, the frequency of a reproduced sound becomes proportionally higher, and in the case of audio or the like, it becomes difficult to hear and it is extremely difficult to grasp the contents. . Conversely, if the playback speed is reduced, the frequency of the playback sound decreases,
A similar thing happens.

On the other hand, in recent years, as shown in FIG. 4, when the reproduction speed is higher than usual, the audio data is thinned out, and when the reproduction speed is lower, the audio data is interpolated. By smoothing the audio data based on the technology, the pitch adjustment technology has been established to keep the frequency of the playback sound constant (constant pitch) even if the playback speed is changed up and down, and to obtain sufficient voice quality for listening. ing.

That is, in accordance with the value of the ratio of the standard speed to the change speed (reproduction speed ratio), the sound data is decimated, interpolated, or adjusted in accordance with each standard pitch, that is, the pitch (frequency) by changing the reproduction speed. A predetermined adjustment process is performed to keep the change of the constant at a constant standard value. Specifically, if the playback speed is faster than the standard speed, the thinning adjustment process is performed, if the playback speed is slower, the interpolation adjustment process is performed, and if the speed is not changed, the playback speed ratio is “1” as described above. The standard speed processing is performed as it is.

[0005] These techniques are described in, for example, Japanese Patent Laid-Open No.
As described in detail in Japanese Patent No. 342763 and JP-A-6-202692, the upper and lower limits of the reproduction speed at this time are set to analog or predetermined fixed values by an external knob signal.

[0006] However, for example, in a place for language training such as conversation (reading based on a script), in order to express the degree of progress, a beginner is 120 words / minute and an intermediate is 15 words / minute.
It is desirable to express the conversation ability digitally with the number of words heard per unit time, such as 0 words / minute and 180 words / minute for advanced users.

[0007]

As described above, the conventional speed control in sound reproduction is analog and intuitive by an external knob signal, and is used in language training such as conversation. There is a problem that digital reproduction speed cannot be set by specifying the number of words per unit time.

In view of the above, the present invention provides means for setting a reproduction speed by specifying the number of words per unit time in a voice reading apparatus in which the pitch is kept constant by increasing and decreasing the reproduction speed. The purpose is to do.

[0009]

SUMMARY OF THE INVENTION The above-mentioned object is achieved by a voice reading apparatus constructed as follows.

FIG. 1 is a block diagram of the present invention. That is, in a voice reading apparatus provided with a pitch adjusting means 15 for changing the reproduction speed, a voice data storage unit 10 for digitally storing voice data, a voice data reproduction speed is set to a number of words corresponding to the voice data and a standard reproduction speed. Standard speed data storage unit 11 for storing standard speed data expressed by time
And a reproduction speed input means 12 for providing reproduction speed change information in terms of the number of words per unit time, and a reproduction speed for obtaining a reproduction speed ratio from the number of words per unit time given by the reproduction speed input means and the number of words at the standard reproduction speed. Ratio calculation means 13
And a control means 14 for outputting the audio data and the standard speed data read from each of the storage units and the reproduction speed ratio to the pitch adjustment means, so that a change in the pitch due to an increase or decrease in the reproduction speed can be maintained at a certain standard. The reproduction speed can be set by specifying the number of words per unit time while keeping the value.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the operation of the configuration diagram of the present invention shown in FIG. 1 will be described with reference to FIG. In step 20 of FIG. 2, the voice data obtained by digitizing the voice data obtained by the reading and the voice data and the standard speed data expressed by the number of words and the standard reproduction time corresponding to the voice data are read into the memory. Then, the ratio of the standard speed to the change speed, that is, the reproduction speed ratio is determined from the number of words at the standard reproduction speed in the read standard speed data and the number of words of the speed change given by the reproduction speed input means.

In step 22, the reproduction speed ratio data is output to the subsequent pitch adjusting means together with the audio data and the standard speed data read from the storage unit.
At this time, if there is no instruction to change the reproduction speed, the reproduction speed ratio is "1" and the reproduction time remains unchanged as the standard reproduction time.

In step 23, according to the value of the reproduction speed ratio, each pitch adjustment process for the audio data is decimated, interpolated, or performed according to the standard, that is, the change of the pitch (frequency) due to the change of the reproduction speed is fixed to a standard value. A predetermined adjustment process is performed to maintain the value. Specifically, if the playback speed is faster than the standard speed, the thinning adjustment process is performed. If the playback speed is slower, the interpolation adjustment process is performed.
If there is no speed change, the playback speed ratio is "1" as described above.
Thus, the standard speed processing is performed as it is.

Then, in step 24, the digital audio data whose pitch has been adjusted is converted into analog audio data (D / A conversion), the audio is reproduced, and one cycle is completed. Next, as an embodiment of the present invention, a speech-to-speech apparatus based on speech data created by reading a language training text of English conversation will be described.

FIG. 3 (1) shows a part of the contents of the teaching materials in this English conversation training.
d restaurant near here? '', `` Please draw a map sho
wing the way to the restaurant. '', `` How long does
It takes from here? ”... and so on. The content of <voice data> in FIG. 3 (2) indicates the data format of voice data in which these conversational sentences are read out and stored as digital voice data, and is composed of voice data and its data length. In addition, silence data exists at a break between conversation sentences. This silence data is of variable length in this example, but this is fixed length,
Silence data can also be represented by a specific mark.

The contents of the <standard speed data> in FIG. 3 (2) indicate the data format of the standard speed data generated and stored in correspondence with the <voice data> at the time of reading out the conversation sentence. Sometimes, it is composed of the standard playback time measured by the built-in timer (when there is no signal for a certain period of time, the silence time is used) and the number of words (the number of words). The sequence number is externally input using the key as a key. At the head of the <standard speed data>, the number of words read per unit time (for example, per minute) is stored as a header as a standard reproduction speed. This standard playback speed is
The sum of each successive standard reproduction time is equal to the unit time (for example, 1
(Minutes), and is an approximate value obtained by selecting an arbitrary area of a continuous sequence No (described later) and counting. Note that a sequence number for synchronizing or managing the contents is attached to both the <voice data> and the <standard speed data>.

[0017] In the content example of the teaching material in FIG.
Is the sequence No. of the content example of the storage unit data in FIG.
For example, in the content example of the teaching material, "Can
The content of "yourecommend a good restaurant near here?" corresponds to the data in the sequence number column of <standard speed data> and <audio data>, and "audio data 1" is "0.6M
The size of "byte" indicates that the standard playback time composed of "8 words" is data of "2.8 seconds". This is the rate at which digital audio data of "0.6 Mbytes" is transmitted in "2.8 seconds". Although there is no description in the content example of the teaching material, or is mainly data indicating a silent state of a break between sentences, the number of words is naturally zero.

Hereinafter, these examples shown in FIG. 3 (standard reproduction speed, 15
The processing procedure will be described based on (0 words / minute). First,
The contents of <standard speed data> and <voice data> shown in FIG. 3 (2) stored in a storage medium such as a disk are read into the memory, and a header of <standard speed data> is read.
It is determined whether there is an instruction to change the standard playback speed in (Header). The instruction to change the reproduction speed is given from the operation unit such as a numeric keypad in the number of words read out per unit time, such as "200" words / minute.

When an instruction to change the playback speed is issued, the playback speed ratio "150/200" = "3/4" is calculated, and the playback speed ratio and the contents of <standard speed data> and <audio data> are calculated. Are transmitted to the pitch adjusting means under the control of the control means in the sequence No. order. (However, the sequence number and word count data are not transmitted.) The pitch adjusting means applies a known pitch adjusting technique, for example, the sequence number “Sound data 1”.
], The standard playback time [2.8] and the playback speed ratio [3]
In the time (2.8 × 3/4 = 2.1 seconds) multiplied by “/ 4”, the data length “0.6 MB” of “audio data 1” is thinned out (0.6 × 3/4) at the playback speed ratio of “3/4”. The pitch adjustment is performed by generating and transmitting “speech data 1” (= 0.45 Mbytes), and a sound at a fixed pitch is reproduced.

As described above, when there is an instruction to change the reproduction speed,
In particular, the playback speed is "2" for the standard playback speed "150" words / minute.
[00] words / min, but when the playback speed is reduced to, for example, [100] words / min,
In the above, the playback speed ratio is set to "150/100" = "3/2", interpolation technology is applied instead of thinning out the audio data, and if there is no instruction to change the standard playback speed, The playback speed ratio is processed as "1".

Such processing is similarly performed continuously for the data after the sequence No. The reproduction speed is maintained until the next reproduction speed change instruction is issued, and the voice reading is performed.

In this embodiment, the reading of the language training textbook for English conversation is described, but the language of the reading is not asked.

[0023]

As is clear from the above description, according to the present invention, it is possible to control the reproduction speed of the reading aloud by externally designating the number of words per unit time, and to freely use digital expressions according to the intention of the user. Speed control
For example, in a place of language training such as conversation, there is an industrial effect that it is possible to objectively evaluate a user's current level and a learning effect digitally.

[Brief description of the drawings]

FIG. 1 is a block diagram of the present invention.

FIG. 2 is an operation flow of the present invention.

FIG. 3 shows an example of contents of storage unit data of the present invention.

FIG. 4 is an explanatory diagram of speed control of conventional audio reproduction.

[Explanation of symbols]

 10 Audio data storage unit 11 Standard speed data storage unit 12 Playback speed input unit 13 Playback speed ratio calculation unit 14 Control unit 15 Pitch adjustment unit 16 Audio playback unit

Continuation of the front page (56) References JP-A-5-342276 (JP, A) JP-A-6-2022692 (JP, A) JP-A 8-202259 (JP, A) JP-A-63-191372 (JP) JP-A-6-149283 (JP, A) JP-A-8-6592 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G10L 21/04

Claims (3)

(57) [Claims] An audio data storage device for digitally storing audio data, comprising: an audio data storage section for digitally storing audio data; and a reproduction speed of the audio data, the number of words corresponding to the audio data and a standard number. A standard speed data storage unit for storing standard speed data expressed by a reproduction time; a reproduction speed input means for giving reproduction speed change information by the number of words per unit time; and the number of words per unit time given by the reproduction speed input means Playback speed ratio calculating means for calculating a playback speed ratio from the number of words at the standard playback speed, and control means for outputting the audio data and the standard speed data read from each of the storage units and the playback speed ratio to the pitch adjustment means. And a speech reading device. 2. The voice reading apparatus according to claim 1, wherein the standard reproduction speed is calculated as an approximate value by the number of words per unit time from the sum of the number of words corresponding to the audio data and the standard reproduction time. . 3. The apparatus according to claim 1, wherein the standard reproduction time is measured and generated by a built-in timer during recording.