JP2002351485A - Electronic mail reading-aloud device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device which reads electronic mail aloud so that it is easy for a user to listen. SOLUTION: A voice quality parameter generating device generates voice quality parameters from information included in the electronic mail and a voice quality parameter changing device changes the generated voice quality parameters; and the changed voice quality parameters are used for voice synthesis for reading the electronic mail aloud.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for reading out digitized character information, and more particularly, to generating and changing a voice quality parameter for reading out an electronic mail from information contained in an electronic mail, and reading out the electronic mail. And a device for performing the same.

[0002]

2. Description of the Related Art As disclosed in Japanese Patent Application No. 4-54985, there is a device for reading out an e-mail aloud using voice parameters corresponding to the sender of the e-mail or voice parameters included in the e-mail.

[0003]

However, there are various environments around a user who receives an electronic mail and uses the voice reading device, and the voice of the voice reading device is not always easy to hear.

SUMMARY OF THE INVENTION It is a first object of the present invention to provide an apparatus for performing a voice reading operation that is easy for a user to hear in consideration of an environment around a user who uses the voice reading apparatus. Further, the user may want to increase or decrease the reading speed of the e-mail.

Accordingly, it is a second object of the present invention to provide a device which enables a user to operate a voice quality parameter so that the user can read aloud a voice at the convenience of the user.

[0006]

Means for Solving the Problems and Effects of the Invention The present invention has the following features in order to achieve the above object. A first invention is an e-mail reading apparatus that performs voice synthesis using a voice parameter and reads an e-mail, and generates an e-mail reading voice parameter from information included in the e-mail. A voice quality parameter changing device for changing a voice quality parameter generated by the voice quality parameter generating device, and an e-mail voice synthesizing device for performing voice synthesis using the voice quality parameter and reading out the e-mail.

According to the first aspect of the present invention, the voice quality parameter generating device generates voice quality parameters from the information included in the electronic mail, and the generated voice quality parameters are changed by the voice quality parameter changing device. Can be used to perform voice synthesis for reading out an e-mail. As a result, it is possible to perform voice reading which is easy for the user to hear.

[0008] A second invention is an invention according to the first invention, wherein the voice parameter changing device changes the voice parameter according to the environment of the user who listens to the sound generated by the electronic mail reading device.

According to the second aspect of the invention, since the environment of the user who listens to the sound generated by the electronic mail reading device can be considered, it is possible to perform the voice reading which is easy for the user to hear. .

[0010] A third invention is an invention according to the second invention, wherein the voice quality parameter changing device uses a voice quality parameter based on information on a vehicle state in which a user is listening to the sound of the e-mail reading device. To change.

[0011] According to the third aspect of the present invention, since the state of the vehicle in which the user who hears the sound of the electronic mail reading device is occupied can be taken into consideration, the user can easily read the voice. Will be able to

A fourth invention is an invention according to the first invention, wherein the voice parameter changing device changes the voice parameter in accordance with a mode set by a user.

[0013] According to the fourth aspect, the voice quality parameter is changed in accordance with the mode set by the user using the electronic mail reading device, and the voice is read out. Can read aloud.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing the structure of an electronic mail reading-out apparatus according to an embodiment of the present invention. The electronic mail reading device 0 includes a voice quality parameter generating device 4, a reading mode setting device 5, a voice quality parameter changing device 6, and an electronic mail voice synthesizing device 7. The e-mail reading device 0 reads the e-mail 1 for reading from the outside, and the reading mode setting device 5 allows the user to read the e-mail 1 so that the user can set the reading method of the e-mail according to the user's convenience. Can be set, and the voice quality parameter can be changed by the voice quality parameter changing device 6 using the vehicle information 3 read from the outside so that the user can easily hear the voice.

[0015] The voice quality parameter generation device 4 generates voice quality parameters (for example, Japanese Patent Application No. 4-549).
A vowel segment, a basic pitch, and a sound source residual signal of each vowel as shown in FIG. 85 are generated from the information included in the electronic mail.

As a method of generating voice quality parameters, there is a method in which a voice quality parameter generation device holds several voice quality parameters and outputs voice quality parameters corresponding to a sender address included in an electronic mail.

If the voice quality parameter itself is included in the electronic mail, there is a method of extracting the voice quality parameter from the electronic mail and outputting the voice quality parameter.
Note that voice parameters can be included in the e-mail by using the e-mail in the MIME format.

The reading mode setting device 5 provides the user with a function that can set the reading mode according to the convenience and preference of the user who uses the electronic mail reading device 0. Functions that can be set by the user include “speech speed”, “speech by a male voice”, and “speech by a female voice”. The setting function of “reading speed” makes it possible to read out quickly or slowly.

FIG. 2 is a block diagram showing the configuration of the voice quality parameter changing device 6. As shown in FIG. Voice quality parameter changing device 6
Comprises a reading parameter reflecting device 61 for changing a voice parameter based on a reading mode parameter output from a reading mode setting device, and a vehicle information reflecting device 62 for changing a voice parameter based on vehicle information.

The reading parameter reflecting device 61 reflects the "reading speed" by changing the length of a phoneme segment included in the voice quality parameter according to the "reading speed" of the reading parameter. More specifically, in the case of "quickly" reading out, the back of the phoneme segment is deleted to shorten the phoneme segment. In speech synthesis, rows of shortened phoneme segments are connected, and as a result, a synthesized speech is read out quickly. If you want to read “slowly,”
By repeating a part of the latter half of the phoneme piece, the length of the phoneme piece is lengthened. In speech synthesis, long rows of phoneme segments are connected, and as a result, speech synthesis speech is read out slowly. FIG. 3A shows an example in which the phoneme segment is shortened by truncating the latter half of the phoneme segment waveform. The reading parameter reflecting device 6 according to this example
1 changes the waveform of the phoneme itself, but analyzes the frequency spectrum of the phoneme in the time axis direction using FFT, wavelet transform, etc., and expands and contracts the length of the frequency spectrum in the time axis direction. The length of the piece may be changed. In this case, after changing the frequency spectrum of the phoneme segments, inverse FFT or inverse wavelet transform is performed to obtain phoneme segments having changed lengths.

A reading parameter reflecting device 61
Shifts the frequency spectrum to the voice quality parameter, thereby reflecting the read-out parameters such as “reading out by male voice” and “reading out by female voice”. FIG. 3 (b) shows that "speaking by a male voice" is realized by shifting the frequency spectrum of a phoneme to a low frequency, and "female voice" is realized by shifting the frequency spectrum of a phoneme to a high frequency. This is an example of “reading by”.

As described above, the reading parameter reflecting device 61 reflects the reading mode set by the user on the speech synthesis by changing the voice quality parameter in accordance with the reading parameter set by the reading parameter setting device 5. In addition, depending on the content of the reading parameter,
There may be a case where the voice quality parameter generated by the voice quality parameter generation device 4 is ignored and the voice quality parameter is changed to a completely different voice quality parameter.

The vehicle information reflecting device 62 changes voice quality parameters that are easy for the user to hear based on the vehicle information. As the vehicle information, “engine speed”, “vehicle speed”, “audio output”, and “road noise” can be considered.

The vehicle information reflecting device 62 changes voice quality parameters that are easy for the user to hear based on the "engine speed". "Engine noise" is generated from the engine. If the frequency spectrum distribution of the "engine noise" with respect to the engine speed is known in advance, the frequency spectrum of the phoneme can be further emphasized so that the phoneme is not buried in the "engine noise". FIG. 4 is a frequency spectrum of a phoneme segment obtained by emphasizing the phoneme segment from the frequency spectrum of the engine noise with respect to the engine speed so that the phoneme segment is not buried in the engine noise.

If the "frequency spectrum distribution of the engine noise with respect to the engine speed" does not change much, the "frequency spectrum distribution of the engine noise with respect to the engine speed" is stored in advance in the vehicle information reflecting device 62. Is also good.

The vehicle information reflecting device 62 estimates "road noise" corresponding to the vehicle speed. Based on the estimated “road noise”, the voice quality parameter is changed to a user-friendly voice quality parameter.

"Road noise" varies depending on the speed of the vehicle, the pattern carved on the tires, and the like. Therefore, "road noise" is estimated by removing components other than "road noise" from the sound obtained by collecting sounds heard in the vehicle with a microphone or the like. FIG. 5 is an example of obtaining the frequency spectrum of “road noise” obtained when “engine noise” and “audio output” are considered as components other than “road noise”. Since the vehicle information includes the vehicle speed, the frequency spectrum of “road noise” can be estimated for various vehicle speeds. From the `` frequency spectrum of road noise with respect to vehicle speed '' obtained in this way, as in the case of changing the voice quality parameter with respect to engine noise, emphasize the phoneme segments so that the phoneme segments are not buried in the road noise. Can change the voice quality parameter.

In order to reduce the influence of the voice of the person riding the vehicle in estimating the "road noise", a sound collecting microphone may be installed outside the vehicle such as a trunk.

The e-mail speech synthesizing device 7 performs a speech synthesizing process on the e-mail read by the e-mail reading device 0, and generates the sound by a speaker 8. At this time,
Voice synthesis is performed based on voice quality parameters output from the voice quality parameter changing device 6. The voice parameter changed by the voice parameter changing device 6 based on the voice parameter generated by the voice parameter generating device from the information included in the e-mail is sent to the e-mail speech synthesizer 7.
Is input to Therefore, the input voice quality parameter is changed according to the information of the vehicle on board so that the reading mode set by the user is reflected and the voice is easy for the user to hear.

It is to be noted that the present invention can provide the same effect not only for the voice reading of electronic mail exchanged on the Internet but also for the reading of digitized text.

Also, in the present invention, as a method of changing the frequency spectrum, FFT or wavelet transform, and inverse FFT
Although the inverse wavelet transform has been increased, it can be implemented by a method of changing the gains of a plurality of band-pass filters constituting the equalizer in order to achieve the simpler implementation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an electronic mail reading device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a more detailed configuration of a voice quality parameter changing device.

FIG. 3 is a diagram showing an example in which a reading parameter reflecting device 61 reflects reading parameters on voice quality parameters;

FIG. 4 is a diagram showing an example in which the vehicle information reflecting device 62 changes voice quality parameters based on “engine speed” that is vehicle information.

FIG. 5 is a diagram showing an example in which the vehicle information reflecting device 62 estimates “road noise” based on vehicle information.

[Explanation of symbols]

 0 ... E-mail reading device 1 ... E-mail 2 ... User 3 ... Vehicle information 4 ... Voice quality parameter generation device 5 ... Reading mode setting device 6 ... Voice quality parameter changing device 61 ... Reading parameter reflecting device 62 ... Vehicle information reflecting device 7 ... E-mail voice synthesizer 8 ... Speaker

Continued on the front page (72) Inventor Takeshi Hatakeyama 1006 Kadoma Kadoma, Osaka Prefecture Inside Matsushita Electric Industrial Co., Ltd. Person Takeshi Kin bronze 1006 Kadoma Kadoma, Kadoma-shi, Osaka Matsushita Electric Industrial Co., Ltd. F term (reference) 5D045 AA07 AB26

Claims (4)

[Claims] 1. A voice quality parameter generating device for generating a voice quality parameter for reading out an email from information included in an email, a voice quality parameter changing device for changing a voice quality parameter generated by the voice quality parameter generating device, and a voice quality parameter changing device. An e-mail reading apparatus including an e-mail speech synthesizing apparatus that performs voice synthesis using the apparatus and reads out an e-mail. 2. The voice quality parameter changing device,
The electronic mail reading device according to claim 1, wherein a voice quality parameter is changed according to an environment of a user who listens to a sound uttered by the electronic mail reading device according to claim 1. 3. The voice quality parameter changing device,
3. The electronic mail reading device according to claim 2, wherein the voice quality parameter is changed according to information on a state of a vehicle in a vehicle, which is an environment of a user who listens to a sound uttered by the electronic mail reading device according to claim 2. 4. The voice quality parameter changing device,
2. The electronic mail reading device according to claim 1, wherein the voice quality parameter is changed according to a mode set by the user.