JPS63225299A - Voice input/output system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an audio input/output system, and particularly to an audio input/output system in information equipment that controls man-machine communication such as personal computers and workstations.

[Conventional technology]

Audio input/output of personal computers, workstations, etc. is performed by direct memory access from or to the main memory (hereinafter referred to as "main memory") in a temporary storage section (as shown in Japanese Patent Application Laid-open No. 159796/1983).
The audio information is once relayed in a buffer memory (hereinafter referred to as buffer memory), and the audio is played back or recorded.

That is, the audio memory consists of a main memory that constantly stores audio information and a buffer memory that temporarily stores audio information during playback and recording.

[Problem that the invention seeks to solve]

In the conventional example described above, since the main memory is accessed each time audio is played back or recorded, there is a problem in that the original data processing performance of the personal computer or workstation is degraded.

An object of the present invention is to provide an audio input/output method that does not reduce the processing performance of the main function of a device due to audio processing.

[Means for solving problems]

The present invention solves the above problem by configuring the buffer memory into a resident part and a non-resident part for audio information.The part that is frequently used during audio output is the resident part, and the part that is used less frequently is the non-resident part. This is achieved by taking it out, storing it, and outputting the audio from there.

[Effect]

Among voice information, voice information such as fixed voice phrases such as guidance and frequently used output voice is transferred from the auxiliary storage device or main memory to the resident part of the buffer memory during initial processing such as when turning on the power of the device. The audio information in the buffer memory is transferred and held, and thereafter the audio information in the buffer memory is used for playback and output until the device is stopped, such as when the power is turned off, and the auxiliary storage device or main memory is not accessed.

〔Example〕

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

In FIG. 1, in a personal computer, a main memory 3 is connected to a system bus 2 of a microprocessor 1.
, and keyboard 7, display 8, auxiliary storage device 9
The audio input/output section 1o of the present invention, which is configured through respective control circuits 4.5.6, is connected to the system bus 2 of the personal computer through an audio connection circuit 11. The audio input/output unit 10 includes a microprocessor 12 and the microprocessor 1 in addition to the audio connection circuit 11.
2, a buffer memory resident section 13a, a non-resident section 13b, a speech synthesis analysis circuit 14, and an A/D and D/A conversion circuit 15 are controlled by the buffer memory resident section 13a.

Audio information to be recorded or registered is inputted from the microphone 17, digitized by the A/D and D/A conversion circuit 15, 2-analyzed and compressed by the speech synthesis analysis circuit 14, and temporarily stored in the buffer memory non-resident section by the microprocessor 12. 13b. The processing of the microprocessors 1 and 12 is synchronized via the audio connection circuit 11, and audio information is transferred to the buffer memory non-resident section 13 using a direct memory access method.
b, and is transferred and stored in the main memory 3 or auxiliary storage device [9]. The storage address at this time is logically designated by input information from the keyboard 7 or cursor position information from the display 8. This audio input processing is the same as before, but since the personal computer itself is intensively executing only this input processing at the time of input, its processing performance is not a problem.On the other hand, while the personal computer is in use, that is, during the period of audio output When audio is output in , the audio output processing occurs along with other main processing, so reading and transferring audio information from the main memory 3 or the auxiliary storage device 9 greatly affects the operating time of the main processing. Standard audio information such as machine guidance and frequently output audio information are always stored in the main memory 3.
Or, it is stored in the auxiliary storage device 9, but when the power of the personal computer is initially turned on, it is automatically transferred from the main memory 3 or the auxiliary storage device i19 to the buffer memory resident section 13a and held there. From now on, when outputting playback,
Upon receiving the small information of the voice configuration code from the host microprocessor 1, the microprocessor 12 takes out the voice information from the buffer memory resident section 13a based on this code information, and converts the voice synthesis and analysis circuit 14 to A/D and D/A. Audio is output to the speaker 16 via the circuit 15. Those that are used less frequently are stored in the buffer memory non-resident section 13b, from which voice information is extracted and outputted as voice to the speaker 16 via the voice synthesis analysis circuit 14, A/D, D/A conversion circuit 15.

The audio information stored in the buffer memory resident section 1.3a and non-resident section 13b disappears when the power of the personal computer is turned off, and the stored information is retrieved from the main memory 3 or the auxiliary storage device 9 again when the power is initially turned on. It will be re-memorized.

Therefore, even if the audio information is frequently used, it is not necessary to access the main memory 3 or the auxiliary storage device 9 each time it is output.

〔Effect of the invention〕

According to the present invention, the processing of the microprocessor 1 of a personal computer or the like is not burdened with the reproduction and output of fixed-form speech sentences and speech sentences with high output frequency, so that the processing efficiency of the personal computer or the like is not reduced. There is no.

[Brief explanation of drawings]

FIG. 1 is a circuit diagram of a personal computer for explaining the operation of an embodiment of the present invention. DESCRIPTION OF SYMBOLS 1...Microprocessor, 2...Microprocessor system bus, 3...Main memory, 4...Keyboard control circuit, 5...Display control circuit, 6...
・Auxiliary storage device control circuit, 7...Keyboard, 8...
Display, 9... Auxiliary storage device, 10... Audio input/output unit, 11... Audio connection circuit, 12... Microprocessor, 13... Buffer memory, 13a.
... Buffer memory resident section, 13b... Buffer memory non-resident section, 14... Speech analysis and synthesis circuit, 15...
A/D, D/A conversion dormitory 1 country

Claims (1)

[Claims] 1. Digitize the audio and store it in either the main storage device or the auxiliary storage device via the temporary storage section, and then transfer the stored information from either the main storage device or the auxiliary storage device to the temporary storage section. In the device that extracts audio, converts it into analog, and outputs the audio, the temporary storage section is composed of a resident section and a non-resident section, and the audio is stored by relaying the non-resident section, and the audio is retrieved through the resident section and the non-resident section. Relay and do it,
During the period of audio output, the stored information is extracted from either the main storage device or the auxiliary storage device and stored in the above-mentioned resident section for frequently used information, and in the above-mentioned non-resident section for infrequently used information. an audio input/output method for outputting audio from the resident section and the non-resident section.