JPH08146985A - Speaking speed control system - Google Patents

Abstract

PURPOSE: To provide a speaking speed control system which can control the speaking speed on the basis of speaking speed control information for a reproducing device which previously stores information for controlling the speaking speed in sent data or on the sound recording media, etc., and receives and reproduces the transmitted data or a reproducing device for the sound recording media. CONSTITUTION: The speaking speed control system is equipped with an editing device 1 which generates a signal by adding speaking speed control information to a speech signal and a reproducing device 2 which separates the signal generated by the editing device 1 into the speech signal and speaking speed control information and performs speaking speed control over the speech signal according to the speaking speed control information.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention includes control information for controlling the speech speed in transmission data such as broadcasting, telephone communication, personal computer, voice mail, etc., and the receiving side talks based on the control information. Speech speed conversion control system to control the speed,
Or recording tape, magneto-optical disk, laser disk,
The present invention relates to a voice speed control system in which control information for controlling the voice speed is recorded in a recording medium such as a CD-ROM, a video CD, an IC memory, etc., and the voice speed is controlled based on the control information during reproduction.

[0002]

2. Description of the Related Art The applicant of the present invention has proposed a speech speed conversion apparatus having means for compressing / expanding the time length of an input voice signal in a voice section and deleting means for deleting an input voice signal in a silent section of a predetermined length or more. We are developing. In such a speech speed conversion device, an input voice signal is analyzed, and compression / expansion processing, deletion processing and the like are performed according to the analysis result.

Input voice signals of such a speech speed converter include broadcast, telephone communication, personal computer, transmission data such as voice mail, recording tape, magneto-optical disk, laser disk, CD-ROM, video CD, There is data read from a recording medium such as an IC memory.

[0004]

DISCLOSURE OF THE INVENTION The present invention provides a reproducing apparatus or a recording medium reproducing apparatus which previously stores information for controlling the speech speed in transmission data, recording medium, etc., and receives and reproduces the transmission data. In order to provide a speech speed control system capable of controlling the speech speed based on the speech speed control information.

[0005]

SUMMARY OF THE INVENTION A speech speed control system according to the present invention includes an editing apparatus for generating a signal in which speech speed control information is added to a speech signal, and a speech signal and a speech speed from a signal generated by the editing apparatus. It is characterized in that it is provided with a reproducing device for separating the control information and controlling the voice speed of the voice signal according to the voice speed control information.

As the editing device, for example, a device for generating a signal in which voice speed control information is added to an audio signal and recording it on a recording medium is used, and as a reproducing device, for example, the voice speed from the recording medium is used. A signal is used in which a signal in which control information is added to a voice signal is read to separate the voice signal from the voice speed control information, and the voice speed of the voice signal is controlled according to the voice speed control information.

As the editing device, for example, a device for generating transmission data by generating a signal in which speech speed control information is added to an audio signal is used, and a reproducing device is as follows.
For example, one that receives the transmission data, separates the voice signal and the voice speed control information from the received transmission data, and controls the voice speed of the voice signal according to the voice speed control information is used.

As a method of adding the voice speed control information to the voice signal, a time division multiplexing system, a frequency division multiplexing system, a switching transmission system, etc. are used.

As the reproducing apparatus, for example, one provided with a speech speed conversion unit having a processing means for time-axis compression / expansion processing of an audio signal and a processing means for deleting the audio signal is used. As the speech speed control information, for example, information for performing a time axis expansion / compression process on an audio signal and information for deleting the audio signal are used. Also, as the voice speed control information, information indicating that it is a voice section,
Information indicating the number of continuous silent sections is used.

An editing apparatus according to the present invention is characterized by including means for analyzing a voice signal to generate voice speed control information, and means for adding the generated voice speed control information to the voice signal.

The reproducing apparatus according to the present invention comprises means for separating the voice signal and the voice speed control information from the voice signal to which the voice speed control information is added, and means for controlling the voice speed of the voice signal according to the voice speed control information. It is characterized by having.

[0012]

In the speech speed control system according to the present invention, the editing device generates a signal in which the speech speed control information is added to the voice signal. Then, on the reproducing device side, the voice signal and the voice speed control information are separated from the signal generated by the editing device, and the voice speed of the voice signal is controlled according to the voice speed control information.

In the editing device according to the present invention, the voice signal is analyzed to generate the voice speed control information, and the generated voice speed control information is added to the voice signal.

In the reproducing apparatus according to the present invention, the voice signal and the voice speed control information are separated from the voice signal to which the voice speed control information is added, and the voice speed of the voice signal is controlled according to the voice speed control information.

[0015]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a speech speed control system.
The speech speed control system includes an editing apparatus 1 that generates a signal in which speech speed control information is added to a voice signal, a voice signal and a speech speed control information that are separated from the signal generated by the editing apparatus 1, and a voice signal. And a reproducing device 2 for controlling the voice speed according to the voice speed control information.

The editing device 1 creates transmission data from a signal in which speech speed control information is added to a voice signal,
There is one in which a signal in which the voice speed control information is added to a voice signal is recorded in the recording medium 3. In the former case, the editing device 1 transmits the transmission data in which the voice speed control information is added to the audio signal to the reproducing device. In the latter case, the editing device 1
Thus, a signal in which the voice speed control information is added to the audio signal is read from the recording medium 3.

FIG. 2 shows the structure of the editing apparatus 1.
In the editing apparatus 1, the original audio signal is the audio signal analysis unit 1
1 to analyze the speech speed control information. The generated speech speed control information is coded by the coding unit 12. Then, the original audio signal and the generated code are multiplexed by the multiplexing unit 13. Transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium.

FIG. 3 shows the structure of the reproducing apparatus 2.
The reproduction device 2 receives the transmission data created by the editing device 1 or the read data from the recording medium on which the multiplexed signal is recorded by the editing device 1.

The input multiplexed signal is demodulated by the demodulator 21 and separated into speech speed control information and a voice signal. The voice signal is input to the speech speed control unit 22. Speech speed control unit 22
The voice speed control information and the voice / speech speed control information synchronizing control signal from the demodulation unit 21 are input to the control signal. The voice speed control unit 22 controls the voice speed of the input voice based on the voice speed control information.

A frequency division multiplexing method, a time division multiplexing method, or the like is used as the multiplexing method. The editing device 1
In, when the voice signal is encoded, the voice signal separated by the demodulation unit 21 is sent to the voice speed control unit 22 after being decoded.

FIG. 4 shows the structure of the speech speed control section 22. The voice speed control unit 22 controls the voice speed control information analysis unit 31 for analyzing the voice speed control information, and the control information for transmitting the synchronization signal to the voice speed control information analysis unit 31 based on the voice / voice speed control information synchronization control signal. A speech speed conversion unit 33 for converting the speech speed of the voice signal based on the analysis results of the synchronization unit 32 and the speech speed control information analysis unit 31 is provided.

The speech speed conversion unit 33 includes a speech speed control information analysis unit 3
According to the analysis result of 1, the compression / expansion processing of the input audio signal
The input voice signal is deleted.

FIG. 5 shows a first operation example of the editing apparatus 1. Here, it is assumed that an audio signal of a broadcast program including video and audio is input. Also, for convenience of explanation,
It is assumed that there are two types of playback speeds set on the playback device 2 side: 1 × speed playback and 2 × speed playback. Further, the voice speed control information multiplexed with the audio signal includes one for 1 × speed reproduction and one for 2 × speed reproduction.

First, the voice signal analyzer 11 calculates the power average value P of the input voice signal for each predetermined section of the input voice signal (step 1). Next, it is determined whether or not the power average value P is equal to or larger than a predetermined threshold Th (step 2). The power average value P is equal to or more than a predetermined threshold Th (P ≧
If it is Th), the section is determined to be a voice section, and is determined to be a section for extending the time length of the voice signal (step 3). Here, the threshold value is set so that low-level stationary noise and environmental sounds other than the voice section are also treated as the silent section. Thereafter, the compression rate α is set (step 4).

The compression rate α for double speed reproduction is, for example, 1
It is set to a predetermined value within the range of / 2≤α≤1. When the compression rate α is 1/2 which is a general compression rate during double speed reproduction, the output audio speed is twice the input audio speed.
When the compression rate α is 1, the output voice speed is 1 time the input voice speed. Therefore, within the range in which the output voice speed is not less than 1 time and not more than 2 times the input voice speed, the compression rate α
Is set.

Here, it is assumed that the compression rate α is set to 2/3. In this case, at the time of reproduction, the 3-pitch cycle of the sound is thinned out to the 2-pitch cycle. Therefore, during reproduction, the output voice speed is 3/2 times the input voice speed. In this way, when compressed at a compression rate of ⅔, only 2 / 3−1 / 2 = 1/6 of the audio is compared with the case of a general compression rate of ½ at the time of double speed reproduction. Will be extended.

The compression rate α for 1 × speed reproduction is, for example, 1
It is set to a predetermined value within the range of ≦ α ≦ 3/2. When the compression rate α is 1 during the 1 × speed reproduction, the output voice speed is 1 × the input voice speed. When the compression rate α is 3/2, the output voice speed is 2/3 times the input voice speed. Therefore, the output voice speed is 2 times the input voice speed.
The compression rate α is set within the range of ⅓ or more and 1 or less.

Here, it is assumed that the compression rate α is set to 3/2. In this case, at the time of reproduction, the 2-pitch cycle of voice is extended to the 3-pitch cycle. Therefore, at the time of reproduction, the output voice speed is ⅔ times the input voice speed. In this way, when the data is expanded at the compression rate of 3/2, it is 3 / 2−1 = 1 as compared with the normal reproduction of the 1 × speed reproduction.
The voice is expanded by / 2.

Next, the voice speed control information having the compression rate α set according to the 1 × speed reproduction speed and the 2 × speed reproduction speed is generated and coded by the coding unit 12 (step 5). Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 6). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium (step 7).

In step 2, the power average value P
Is smaller than a predetermined threshold Th (P <T
h), the section is determined to be a silent section, and the number of continuous silent sections is calculated (step 8). Then, it is determined whether or not the number of continuous silent sections is equal to or larger than the predetermined number Tdel (step 9). The number of continuous silent sections is a predetermined number Tde
When it is less than l, it is determined to be a section in which the time length of the audio signal is extended (step 10).

Next, similarly to step 4, the compression rates α for 1 × speed reproduction and 2 × speed reproduction are set (step 11). Then, the voice speed control information including the compression rate α for the 1 × speed reproduction and the 2 × speed reproduction is generated and encoded by the encoding unit 12 (step 11). next,
The multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 6). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium (step 7).

When it is determined in step 9 that the number of continuous silent sections is greater than or equal to the predetermined number Tdel,
The section is determined to be a section to be deleted (step 1
2). Then, the control information in which the input audio signal of the section is set as the deletion section is generated and coded by the coding unit 12 (step 13).

Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 6). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium (step 7).

When the multiplexed signal thus generated is reproduced on the reproducing device 2 side, the reproduction speed (1 × speed or 2 × speed) is set by the operator. In the section in which the compression rate α for the set playback speed is set as the voice speed control information in the input audio signal, the compression rate α is set.
At, the input audio signal is time-axis compressed / decompressed and then reproduced. Further, in the input voice signal, the input voice signal is deleted in the section designated to be deleted by the voice speed control information.

FIG. 6 shows a second operation example of the editing apparatus 1. Here, it is assumed that an audio signal of a broadcast program including video and audio is input. Also, for convenience of explanation,
It is assumed that there are two types of playback speeds set on the playback device 2 side: 1 × speed playback and 2 × speed playback. Further, the voice speed control information multiplexed with the audio signal includes one for 1 × speed reproduction and one for 2 × speed reproduction.

First, the voice signal analysis unit 11 calculates the power average value P of the input voice signal for each predetermined section of the input voice signal (step 21). Next, it is determined whether or not the power average value P is equal to or larger than a predetermined threshold Th (step 22). The power average value P is equal to or more than a predetermined threshold Th (P
If ≧ Th), the section is determined to be a voice section, and is determined to be a section for extending the time length of the voice signal (step 23). Then, the voice speed control information indicating that the time length of the audio signal is to be extended is generated and encoded by the encoding unit 12 (step 24).

Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 25). Transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium (step 26).

In step 22, when the power average value P is smaller than the predetermined threshold value Th (P <T
h), the section is determined to be a silent section, and the number of continuous silent sections is calculated (step 27). Then, the speech rate control information indicating the number of continuous silent sections is generated by the encoding unit 12 (step 28).

Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 25). Transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on a recording medium (step 26).

When the multiplexed signal thus generated is reproduced on the reproducing device 2 side, the reproduction speed (1 × speed or 2 × speed) is set by the operator. Further, the operator sets a compression rate α according to the set reproduction speed. In addition, the continuation length Tdel for determining the deleted section
Is set by the operator.

Then, of the input voice signal, in the section designated to be expanded by the voice speed control information,
The input audio signal is subjected to time-base compression / expansion processing at the compression rate α set by the operator. Also, of the input audio signal,
The duration length Tdel in which the duration number of the silent section is designated by the speech speed control information and the duration length is determined by the operator.
In the section above, the input audio signal is deleted. In the input voice signal, the compression rate α set by the operator is set in a section in which the number of continuous silent sections is designated by the voice speed control information and the continuous length is less than the continuous length Tdel determined by the operator. At, the input audio signal is subjected to time axis compression / expansion processing.

FIG. 7 shows a third operation example of the editing apparatus 1. Here, it is assumed that an audio signal of a broadcast program including video and audio is input. Also, for convenience of explanation,
It is assumed that there are two types of playback speeds set on the playback device 2 side: 1 × speed playback and 2 × speed playback. Further, the voice speed control information multiplexed with the audio signal includes one for 1 × speed reproduction and one for 2 × speed reproduction.

First, the voice signal analyzer 11 calculates the power average value P of the input voice signal for each predetermined section of the input voice signal (step 31). Next, it is determined whether or not the power average value P is equal to or greater than a predetermined threshold Th (step 32). The power average value P is equal to or more than a predetermined threshold Th (P
≧ Th), the section is determined to be a voice section, and is determined to be a section for extending the time length of the voice signal (step 33).

Next, the amount of audio expansion at the present time, that is, a value corresponding to the delay time of the output signal with respect to the input signal during reproduction is calculated for each of the 1 × speed and the 2 × speed (step 34). ). Then, the compression rate α according to the expansion amount is set for each of the 1 × speed and the 2 × speed (step 35).

The compression rate α for double speed reproduction is, for example, 1
It is set within the range of / 2≤α≤1, and the compression rate α is increased as the expansion amount at the present time is smaller. When the compression rate α is 1/2 which is a general compression rate during double speed reproduction,
The output voice speed is twice the input voice speed, and when the compression rate α is 1, the output voice speed is one time the input voice speed. Therefore, the output voice speed is 1 times the input voice speed.
The compression rate α is set within the range of not less than twice and not more than twice.

Further, the compression rate α is set so that the output voice speed becomes faster as the amount of expansion at the present time becomes larger. The reason for this is to prevent the delay time of the output signal from the input signal from exceeding a predetermined time.

The compression rate α for 1 × speed reproduction is, for example, 1
The compression rate α is set to a predetermined value within the range of ≦ α ≦ 3/2, and the smaller the expansion amount at the present time, the larger the compression rate α. When the compression rate α is 1 during the 1 × speed reproduction, the output voice speed is 1 × the input voice speed. When the compression rate α is 3/2, the output voice speed is 2/3 times the input voice speed. Therefore, the output voice speed is 2 times the input voice speed.
The compression rate α is set within the range of ⅓ or more and 1 or less.

Regarding the compression rate α for 1 × speed reproduction, the compression rate α is set so that the output voice speed becomes faster as the amount of expansion at the present time becomes larger. The reason for this is to prevent the delay time of the output signal from the input signal from exceeding a predetermined time.

Next, the voice speed control information consisting of the compression rate α set according to the 1 × speed reproduction speed and the 2 × speed reproduction speed is generated and coded by the coding unit 12 (step 36). Next, the multiplexing unit 13
The input voice signal of the section and the voice speed control information are multiplexed (step 37). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on the recording medium (step 38).

In step 32, when the power average value P is smaller than the predetermined threshold value Th (P <T
h), the section is determined to be a silent section, and the number of continuous silent sections is calculated (step 39). Then, it is determined whether or not the number of continuous silent sections is equal to or larger than the predetermined number Tdel (step 40). The number of continuous silent sections is a predetermined number T
When it is less than del, it is determined to be a section in which the time length of the audio signal is extended (step 41).

Next, as in step 34, the amount of audio expansion at the present time, that is, the value corresponding to the delay time of the output signal with respect to the input signal at the time of reproduction is set to the 1 × speed and the 2 × speed. Is calculated (step 42).
Then, similarly to step 35, the compression rate α according to the expansion amount is set for the 1 × speed and the 2 × speed (step 43).

Then, the voice speed control information including the compression rate α set according to the 1 × speed reproduction speed and the 2 × speed reproduction speed is generated and encoded by the encoding unit 12 (step 44). Next, the multiplexing unit 13 multiplexes the input voice signal and the voice speed control information in the section (step 37). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on the recording medium (step 38).

When it is determined in step 40 that the number of continuous silent sections is equal to or larger than the predetermined number Tdel, the section is determined to be deleted (step 45). Then, the encoding unit 12 generates control information in which the input audio signal of the section is the deletion section (step 46).

Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 37). Then, transmission data is created based on this multiplexed signal, or this multiplexed signal is recorded on the recording medium (step 38).

When the multiplexed signal thus generated is reproduced on the reproducing apparatus 2 side, the reproducing speed (1 × speed or 2 × speed) is set by the operator. In the section in which the compression rate α for the set playback speed is set as the voice speed control information in the input audio signal, the compression rate α is set.
At, the input audio signal is time-axis compressed / decompressed and then reproduced. Further, in the input voice signal, the input voice signal is deleted in the section designated to be deleted by the voice speed control information.

In the third operation example, as in the second operation example shown in FIG. 6, when it is determined that there is a silent section (NO in step 32), the duration of the silent section is generated as the voice speed control information. You may do it.

FIG. 8 shows a fourth operation example of the editing apparatus 1. Here, it is assumed that an audio signal of a broadcast program including video and audio is input. Also, for convenience of explanation,
It is assumed that there are two types of playback speeds set on the playback device 2 side: 1 × speed playback and 2 × speed playback. Further, the voice speed control information multiplexed with the audio signal includes one for 1 × speed reproduction and one for 2 × speed reproduction.

First, the voice signal analyzer 11 calculates the power average value P of the input voice signal for each predetermined section of the input voice signal (step 51). Next, it is determined whether the power average value P is greater than or equal to a predetermined threshold Th (step 52).

When the power average value P is equal to or larger than the predetermined threshold Th (P ≧ Th), it is determined that the section is a voice section. Then, based on the state of the flag F,
It is determined whether or not the previous section was a silent section whose continuation number is equal to or greater than the predetermined number Tdel (step 53).

When the previous section is not a silent section in which the number of continuations is the predetermined number Tdel or more (F = 0), the compression rate α is set to monotonically decrease with respect to the currently set compression rate. (Step 54). Compression rate α for double speed playback
Is set within the range of ½ ≦ α ≦ 1.
The compression rate α is 1/2 which is a general compression rate during double speed reproduction.
, The output voice speed is twice the input voice speed, and when the compression rate α is 1, the output voice speed is 1 time the input voice speed. Therefore, the compression rate α is set within a range in which the output voice speed is not less than 1 time and not more than 2 times the input voice speed.

In step 54, the compression rate α is monotonically decreased, which means that the output voice speed during double-speed reproduction is not less than 1 time and not more than twice the input voice speed.
This means that the compression rate α is set so that the output audio speed during reproduction becomes faster. This is to prevent the delay time of the output audio signal with respect to the input audio signal from exceeding a predetermined time during double speed reproduction.

The compression rate α for 1 × speed reproduction is, for example, 1
It is set within the range of ≦ α ≦ 3/2. When the compression rate α is 1, the output voice speed is 1 time the input voice speed. When the compression rate α is 3/2, the output voice speed is 2/3 times the input voice speed. Therefore, the compression rate α is set within a range in which the output voice speed is ⅔ or more times the input voice speed and is 1 time or less.

In step 54, the compression rate α is monotonically decreased, meaning that during 1 × speed reproduction, reproduction is performed within a range in which the output audio speed is ⅔ or more times the input audio speed and less than or equal to 1 time. Compression rate α so that the output voice speed at
Means that is set. This is to prevent the delay time of the output audio signal from the input audio signal from exceeding a predetermined time during the 1 × speed reproduction.

Next, the speech speed control information consisting of the compression setting α set in accordance with the 1 × speed reproduction speed and the 2 × speed reproduction speed is generated and coded by the coding unit 12 (step 55). Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 56). Then, transmission data is created based on this multiplexed signal or this multiplexed signal is recorded on the recording medium (step 57).

In step 53, the previous section is a silent section whose continuation number is equal to or more than the predetermined number Tdel (F
= 1), the flag F is reset (F
= 0) (step 58). Then, the compression rate α is set to be larger than the currently set compression rate (step 59). That is, when the compression rate α for 2 × speed reproduction is set within the range of 1/2 ≦ α ≦ 1, the output audio speed is 1 or more times and 2 times or less than the input audio speed during the 2 × speed reproduction. The compression rate α for double speed reproduction is set so that the output audio speed during reproduction becomes slower within the range. Further, the compression rate α for 1 × speed reproduction is 1 ≦ α ≦ 3
When set within the range of / 2, the output voice speed is 1/3 when the output voice speed is ⅔ or more times the input voice speed during 1 × speed reproduction.
The compression rate α for 1 × speed reproduction is set so that the output audio speed during reproduction becomes slow within the range of double or less.

Then, the voice speed control information including the compression setting α set according to the 1 × speed reproduction speed and the 2 × speed reproduction speed is generated and coded by the coding unit 12 (step 55). Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 56). Then, transmission data is created based on this multiplexed signal or this multiplexed signal is recorded on the recording medium (step 57).

In step 52, when the power average value P is smaller than the predetermined threshold value Th (P <T
h), the section is determined to be a silent section, and the number of continuous silent sections is calculated (step 60). Then, it is determined whether or not the number of continuous silent sections is equal to or larger than the predetermined number Tdel (step 61). The number of continuous silent sections is a predetermined number T
When it is less than del, the flag F is set (F =
1) (step 62). Also, it is determined that the time length of the audio signal is to be extended (step 63).

Next, as in step 54, the compression rates α for 1 × speed reproduction and 2 × speed reproduction are set so as to monotonically decrease with respect to the currently set corresponding compression rates (step 64). ). The compression setting α set according to the 1 × speed reproduction speed and the 2 × speed reproduction speed, respectively.
The voice speed control information consisting of is generated and encoded by the encoding unit 12 (step 65). Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 56). Then, transmission data is created based on this multiplexed signal or this multiplexed signal is recorded on a recording medium (step 5).
7).

When it is determined in step 61 that the number of continuous silent sections is equal to or greater than the predetermined number Tdel, the section is determined to be a section to be deleted (step 66). Then, the control information in which the input voice signal of the section is set as the deletion section is generated and coded by the coding unit 12 (step 67).

Next, the multiplexer 13 multiplexes the input voice signal and the voice speed control information in the section (step 56). Then, transmission data is created based on this multiplexed signal or this multiplexed signal is recorded on the recording medium (step 57).

When the multiplexed signal thus generated is reproduced on the reproducing apparatus 2 side, the reproducing speed (1 × speed or 2 × speed) is set by the operator. In the section in which the compression rate α for the set playback speed is set as the voice speed control information in the input audio signal, the compression rate α is set.
At, the input audio signal is time-axis compressed / decompressed and then reproduced. Further, in the input voice signal, the input voice signal is deleted in the section designated to be deleted by the voice speed control information.

In this fourth operation example as well, as in the second operation example of FIG. 6, when it is determined to be a silent section (NO in step 52), the duration of the silent section is generated as the voice speed control information. You may do it.

Further, as shown in FIG. 9, the length of one sentence of a voice sandwiched between a predetermined number or more of silent sections is measured, and the compression rate α is monotonically decreased according to the length of one sentence. May be. That is, the speech rate control information may be created such that the compression rate α increases at the beginning of a sentence (output voice speed becomes slower), and the compression rate α becomes smaller at the end of a sentence (output voice speed increases).

At a point where the pitch of the input voice sharply rises, the compression rate α becomes large (the output voice speed becomes slow), and thereafter, as the pitch of the input voice falls,
So that the compression rate α is small (the output voice speed is high),
The speech speed control information may be created.

Further, the utterance speed of the input voice may be detected and the compression rate α may be determined according to the utterance speed. That is, when the utterance speed is high, the compression rate α is high (the output voice speed is low), and when the utterance speed is low, the compression rate α is low (the output voice speed is high). Control information may be created. As a method of detecting the utterance speed, there are a method of detecting the number of vowels per unit time or a duration of vowels, a method of detecting a ratio of a voice section to a silent section per unit time, and the like.

In each of the above embodiments, the input voice is divided into the voice section and the silent section, but the input voice may be divided into the voiced section, the unvoiced section and the silent section. In these determinations, the voiced section is first distinguished from the unvoiced section and the silent section based on the power average value. Next, the unvoiced section and the unvoiced section are distinguished by calculating the number of zero crossings of the voice signal (the number of crossings with the zero level per unit time), and distinguishing the section in which the number of zero crossings is a predetermined number or more as the unvoiced section. , A section in which the number of zero crossings is less than a predetermined number is determined as a silent section. The voice speed control information is created for the voiced section as a section in which the time length of the voice signal is extended, a silent section having a predetermined length or more as a deleted section, and the unvoiced section as an unprocessed section. In the case of a voiced section, the pitch of the voice may be extracted by using the autocorrelation method, and this pitch extraction information may be created as the speech speed control information.

When the audio signal is music, in order to prevent the compression processing from being performed on the reproducing device side,
You may make it insert the information which shows that it is music as speech speed control information. As a method of discriminating whether or not it is music, frequency analysis is performed on the input voice signal, and high frequency range (4
KHz or higher) and low frequency range (less than 4 KHz)
There is a method of calculating the ratio with the signal component of, and judging as music when the high frequency band component is large.

Further, in a technical information program or the like, a phrase of a predetermined technical term is detected by analyzing an audio signal, and a predetermined keyword is added to the detected section as speech speed control information, and at the time of reproduction. When the operator sets the keyword in the reproducing device, the section (phrase of the technical term) corresponding to the keyword may be slowly reproduced.

In the above embodiment, the voice signal and the voice speed control information are multiplexed and output by the editing apparatus. However, the voice signal and the voice speed control information are switched and output by the transmission method. Good.

On the reproducing device side, 1 × speed reproduction,
In addition to the 2 × speed reproduction, it is preferable to set the compression rate α for all reproduction magnification speeds and add it to the voice signal as the voice speed control information even when the 3 × speed reproduction is possible.

[0082]

According to the present invention, information for controlling the speech speed is previously stored in transmission data, recording media, etc., and the speech speed can be controlled on the receiving side or the reproducing side based on the control information. A fast conversion system is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a speech speed control system.

FIG. 2 is a block diagram showing a configuration of an editing device.

FIG. 3 is a block diagram showing a configuration of a playback device.

FIG. 4 is a block diagram showing a configuration of a speech speed control unit of the playback device.

FIG. 5 is a flowchart showing a first operation example of the editing apparatus.

FIG. 6 is a flowchart showing a second operation example of the editing apparatus.

FIG. 7 is a flowchart showing a third operation example of the editing apparatus.

FIG. 8 is a flowchart showing a fourth operation example of the editing apparatus.

FIG. 9 is a schematic diagram showing a section of one sentence sandwiched between silent sections of a predetermined length or more.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Editing device 2 Playback device 11 Voice signal analysis unit 12 Speech rate control information coding unit 13 Multiplexing unit 21 Demodulation unit 22 Speech rate control unit 31 Speech rate control information analysis unit 32 Control information synchronization unit 33 Speech rate conversion unit

Claims (9)

[Claims] 1. An editing device for generating a signal in which voice speed control information is added to a voice signal, and a voice signal and voice speed control information are separated from the signal generated by the edit device, and the voice signal is transmitted at a voice speed. A speech speed control system including a reproducing device for controlling the speech speed according to control information. 2. The editing device generates a signal in which the voice speed control information is added to the voice signal and records the signal on a recording medium, and the reproducing device outputs the voice speed control information from the recording medium to the voice signal. 2. The voice speed control system according to claim 1, wherein the added signal is read to separate the voice signal and the voice speed control information from each other, and the voice speed of the voice signal is controlled according to the voice speed control information. 3. The editing device generates a signal in which the voice speed control information is added to the audio signal to create transmission data, and the reproducing device receives the transmission data and receives the transmission. 2. The voice speed control system according to claim 1, wherein the voice signal and the voice speed control information are separated from the data, and the voice speed of the voice signal is controlled according to the voice speed control information. 4. The voice speed control system according to claim 1, wherein the voice speed control information is added to the voice signal by a time division multiplexing or a frequency division multiplexing method. 5. The voice speed control system according to claim 1, wherein the voice speed control information is added to the voice signal by the switching transmission method. 6. The reproduction device includes a speech speed conversion unit having a processing unit for time-axis compression / expansion processing of the audio signal and a processing unit for processing the deletion of the audio signal, and the speech speed control information is the audio signal. 2. At least information for performing time-axis expansion / compression processing and information for deleting audio signals are included.
6. The speech speed control system according to any one of 2, 3, 4 and 5. 7. The reproducing device comprises a speech speed conversion unit having a processing means for time-axis compression / expansion processing of an audio signal and a processing means for deleting an audio signal, wherein the speech speed control information is a voice section. 6. The speech speed control system according to claim 1, wherein the speech speed control system includes at least information indicating that the number of silent periods is continuous and information indicating the number of continuous silent periods. 8. An editing apparatus comprising means for analyzing a voice signal to generate voice speed control information, and means for adding the generated voice speed control information to the voice signal. 9. A reproducing apparatus comprising means for separating a voice signal and voice speed control information from a voice signal to which voice speed control information is added, and means for controlling the voice speed of the voice signal according to the voice speed control information. ..