JPS62278860A - Voice storage device - Google Patents

Abstract

PURPOSE:To decide the priority sequence of a message stored by the importance degree and emergency degree of the message, to store an answer message to the contents of the message and to communicate this to a speaking person at a transmitting call side by controlling the message stored by a key word from the speaking person. CONSTITUTION:After a network control circuit 1 automatically receives a call transmitted from a speaking person through a subscriber telephone line, a sound responding circuit 5 requests the speaking of a name or other key word with a synthetic sound to the speaking person. In such case, a voice (name or key word) spoken by the speaking person is inputted to a voice storage circuit 4 and a speech recognizing circuit 3. The voice storage circuit 4 accumulates all spoken voice. The speech recognizing circuit 3 reads a pattern registered beforehand from a sound pattern registering circuit 2, recognizes the voice inputted by the pattern and reports the result to a control circuit 6 when the equivalent name or key word exists. The audio responding circuit 5 requests the input of a message to the speaking person, stores the contents to the voice storage circuit 4 wholly and controls in the relation to the name or key word.

Description

[Detailed description of the invention] 3. Detailed description of the invention [Industrial application field] The present invention is suitable for use in an answering machine.

In particular, the present invention relates to a voice storage device that stores and manages messages from a telephone line.

〔overview〕

The present invention facilitates the handling of accumulated messages in a voice storage device that stores messages from a telephone line by accumulating the messages and managing the contents based on keywords from the speaker. .

[Conventional technology]

Traditional answering machines automatically answer incoming telephone calls and send prerecorded or synthetic voice messages.
Nosage can request a message from the caller and store its contents.

[Problem that the invention seeks to solve]

However, when conventional answering machines store messages from multiple callers, they only store the messages in order, but cannot manage their order, callers, or content. There wasn't. Therefore, those who received the message could not know its contents unless they reproduced all the messages.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above problems and provide an audio storage device that can manage stored contents.

[Means for solving problems]

The voice storage device of the present invention is a voice storage device equipped with a voice storage circuit for storing messages from a telephone line, and includes a voice recognition circuit and a means for requesting a keyword from a speaker who transmits the message. The voice recognition circuit is characterized in that it includes means for recognizing a received keyword by comparing it with a pre-registered pattern, and means for managing the contents stored in the voice storage circuit based on the keyword. Here, "keywords" include words for identifying the speaker's name, the content of the message, etc.

[Function] The voice storage device of the present invention manages messages stored from speakers using key words. Therefore, it is possible to prioritize the accumulated messages according to their importance and urgency, and it is possible to accumulate reply messages to the contents of the messages and convey them to the calling party.

The voice storage device of the present invention is suitable for use as an answering machine, but can also be used by connecting to a PBX, exchange, etc.

〔Example〕

FIG. 1 is a block diagram of an audio storage device according to an embodiment of the present invention.

The network control circuit 1 is connected to a voice pattern registration circuit 2, a voice recognition circuit 3, a voice storage circuit 4, and a voice response circuit 5. The voice pattern registration circuit 2, the voice recognition circuit 3, the voice storage circuit 4, and the voice response circuit 5 are connected to and controlled by a control circuit 6.

After the network control circuit 1 automatically receives a call originating from the speaker via the subscriber's telephone line, the voice response circuit 5 requests the speaker to utter his/her name or other key words in a synthesized voice. . For example, the user may be guided by saying, ``Please enter your name, please enter a keyword.'' At this time, the voice (name or keyword) uttered by the speaker is input manually to the voice storage circuit 4 and the voice recognition circuit 3. The voice storage circuit 4 stores all uttered voices. The speech recognition circuit 3 reads a pre-registered pattern from the speech pattern registration circuit 2, recognizes the input speech using this pattern, and reports the result to the control circuit 6 if there is a corresponding name or keyword. . Furthermore, the voice response circuit 5
The speaker is requested to input a message, all the contents are stored in the voice storage circuit 4, and managed in association with the name or keyword. The call is then completed by disconnecting the line.

If the voice input when inputting the name or keyword manually has not yet been registered in the voice pattern registration circuit 2, the speaker cannot be identified, so the voice is managed as a new speaker. In this case, after the line is disconnected, the stored name or keyword voice is transferred from the voice storage circuit 4 to the voice pattern registration circuit 2, and registered as a new pattern by learning. Therefore, the next time there is a call from the same speaker, it becomes possible to identify the speaker.

〔Effect of the invention〕

As explained above, the voice storage device of the present invention can simultaneously input voice into the voice recognition circuit and the voice storage circuit, identify the speaker, and in the case of a new speaker, learn the voice. can be registered as a pattern. Therefore, there is an effect that the accumulated messages and speakers can be managed in association with each other.

When the voice recognition device of the present invention has accumulated a reply message to a message, the next time there is a call from the same speaker, the voice recognition device automatically utters the reply message (
It is possible to play it back and listen to it. In addition, if a display screen or printer is connected, it is possible to create a list based on the speakers and key words of the stored messages, and you can select and listen to any message as needed. Can be done. Furthermore, since the name of the speaker can be recognized after receiving a call, the receiver can answer the call directly if necessary, allowing the receiver to make a free choice depending on the speaker and the content.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an audio storage device according to an embodiment of the present invention. Figure 2 is a flowchart. 1... Network control circuit, 2... Voice pattern registration circuit,
3... Voice recognition circuit, 4... Voice storage circuit, 5...
-Voice response circuit, 6...control circuit.

Claims (1)

[Claims] (1) A voice storage device equipped with a voice storage circuit for storing messages from a telephone line, comprising a voice recognition circuit and means for requesting a keyword from a speaker transmitting the message, the voice recognition circuit comprising: A voice storage device comprising means for recognizing a received keyword by comparing it with a pre-registered pattern, and managing the contents stored in the voice storage circuit based on the keyword.