JPH0731508B2 - Speech recognition response 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 voice recognition response device used in an information processing system by voice input.

[0002]

2. Description of the Related Art Recently, the speech recognition technology and the speech synthesis technology have been remarkably developed. For example, continuous speech recognition and speech recognition for an unspecified speaker are possible, and the accuracy using the linear predictive coding method is improved. High voice synthesis is possible. Also, research and development have been actively conducted on a rule synthesis method for converting sentences into speech.

Using such a technology, for example, a telephone voice response service system for performing various services using a public telephone line and an online business system in a bank or the like have been attempted to be developed, and its usefulness has been noted. ing.

[0004]

However, the number of users of this type of system is unspecified, and for example, there are unfamiliar people such as old people and children, or people who use it many times a day. In spite of this, in the conventional device, the content of the voice response is uniform and the utterance speed is also constant, so that the conversation between the human and the machine is not smooth. In other words, the response is redundant and frustrating,
Or there was a problem that the response was difficult to understand.

The present invention has been made in consideration of such circumstances, and the purpose thereof is to enable a natural and smooth dialogue between a human and a machine and enable effective information processing by voice input. It is to provide a highly practical voice recognition response device that can be used.

[0006]

The present invention, in a voice recognition response device for recognizing an input voice and outputting a response sentence in response to the recognition result, is the same as the measuring means for measuring the utterance speed of the input voice. The present invention is characterized in that an expression format having a different number of characters is prepared in advance for a response sentence representing meaning, and a control means for controlling the expression format of the response sentence according to the utterance speed measured by the measuring means is provided.

[0007]

SUMMARY OF above, in the present invention, in accordance with the speech speed of the input speech, by control the expression format of the response sentence, it is possible to provide an appropriate response to the speech input person.
For example, by giving a simple response with a small number of characters to a frequently used person and a polite response with a large number of characters to an infrequently used person, it is possible to give an appropriate instruction for voice input. It becomes possible and the naturalness and smoothness of the dialogue can be sufficiently enhanced.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of a voice recognition response device according to an embodiment of the present invention. This device detects the utterance speed of the input voice and changes the content itself of the response sentence. That is, the input voice is analyzed via the analyzer 11, and the voice pattern obtained by this analysis is stored in the voice pattern memory 12. The voice recognition unit 13 refers to the voice dictionary registered in the dictionary memory 14 to recognize this voice pattern.

On the other hand, the pitch frequency component of the input voice is extracted through the pitch extractor 15. The extraction of the pitch frequency component is performed independently of the recognition processing of the input voice by using, for example, the cepstrum method, the modified correlation method, the ADM method, or the like. From the change of the time-series pattern of the pitch frequency, the speech rate measuring device 16 obtains the speech rate v of the input voice. Voice response control unit 17
Inputs the information of the speech speed v and the recognition result of the input voice and determines a response sentence having contents corresponding thereto, and this is output as voice through the voice response output unit 18.

That is, depending on the voice recognition result and the speech rate, even if the response has the same meaning such as "Thank you", "Thank you", "Thank you again, please", etc. (word count)
One of the different ones is selectively controlled and output as voice. That is, the voice response having the content and speed according to the voice input person is made. Therefore, when giving the next voice input instruction as a voice response, it is possible to give the instruction simply with a small number of characters, or to give the instruction carefully with a large number of characters to an unfamiliar person, It is possible to enhance the naturalness of the dialogue and improve the processing efficiency.

The present invention is not limited to the above embodiment. For example, the speech rate of the input voice is measured by utilizing the vowel similarity and the temporal change of the spectrum, and the continuous occurrence of numbers is recognized. When the target is set, the pause time length can be used. In addition, the control of the voice response may be performed by controlling the utterance speed together with the control of the content sentence (expression format) . The present invention is needed, it can be modified in various ways without departing from the scope thereof.

[0012]

As described above, according to the present invention, the voice utterance speed that well reflects the character of the voice input person is detected, and the expression form (the number of characters) of the response sentence of the voice response is controlled accordingly. It is possible to enhance the naturalness of the dialogue with the voice input person. As a result, practically great effects such as elimination of troubles such as annoyance to the voice input person can be obtained.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a voice recognition response device according to an embodiment of the present invention.

[Explanation of symbols]

11 ... Analyzer 12 ... Voice pattern memory 13 ... Voice recognition unit 14 ... Dictionary memory 15 ... Pitch extractor 16 ... Speech rate measuring unit 17 ... Voice response control unit 18 ... Voice response output unit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G06F 17/28

Claims (1)

[Claims] 1. A voice recognition response device for recognizing an input voice and outputting a response sentence to the recognition result by voice, comprising: a measuring means for measuring the utterance speed of the input voice; A voice recognition response device, comprising different expression formats prepared in advance, and control means for controlling the expression format of the response sentence in accordance with the speech rate measured by the measurement means.