JP7577700B2 - Program, terminal and method for assisting users who cannot speak during online meetings - Google Patents

Description

The present invention relates to technology for an online conference system in which multiple users participate using terminals.

Online conference systems that allow multiple users to hold conferences simultaneously using terminals such as smartphones and personal computers are now in widespread use. Videos of multiple users are displayed on the terminal's display, the audio of the conference is output from the terminal's speaker, and the users' speech is picked up by the terminal's microphone. Online conference systems also generally have a function that allows users to post text chat.

2. Description of the Related Art Conventionally, there are techniques for achieving smooth communication when a dialogue system that converses with a person speaks by itself.
For example, there is a technique in which a dialogue system analyzes the acoustic and linguistic features of the content of a user's utterance to determine the end state of the utterance and whether or not the user has the authority to speak (see, for example, Patent Document 1). This technique allows the dialogue system to detect the appropriate timing to speak to the user.
There is also a technique in which a dialogue system expresses its intention to speak by making breathing sounds when it speaks, thereby hinting to those around it when it is about to speak (see, for example, Patent Document 2).

JP 2021-051172 A Patent No. 6712303

In an online conference system, members participating in the conference can freely speak and discuss. However, some members may be in an environment where they cannot speak. These members usually end up only listening to the discussions that other members are speaking. In order to make their own opinions known, they must post them in a text chat, for example.

FIG. 1 is a schematic diagram of an online conference system in the prior art.
According to Fig. 1, during an online conference, it is assumed that users @b and @c are in an environment where they can speak, but user @a is in an environment where he cannot speak. User @a may be traveling by train, for example. User @a in such an environment can hear the discussion of the online conference, but it is difficult for him to speak himself.

FIG. 2 is a screen diagram of the online conference system shown in FIG.
According to Fig. 2, speech is exchanged during an online conference as follows: "" is spoken voice data.
User @c: "X company's equipment is cheap."
User @b: "But company Y's management functions are excellent."
User @a "..."
At this time, user @a can write his/her own opinion in the text chat field even during the online conference. {} is the entered chat text.
User @c {Even though it is expensive, considering the maintenance, we should choose company y.}

However, even during online meetings, not all participants are always paying attention to the text chat, so not only will your message go unnoticed, but if the topic quickly changes, you will miss the timing entirely.

In response to this, the inventors of this application wondered if it would be possible to assist users who are unable to speak during online meetings. In particular, they wondered if it would be possible to communicate the opinions of users who are in an environment where they cannot speak to all members as timely as possible.

The present invention aims to provide a program, terminal, and method that can assist users who cannot speak during online conferences.

According to the present invention, there is provided a program for causing a computer installed in a terminal that exchanges voice data in real time via an online conference server to function, the program comprising:
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
causing the computer to function as chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing ;
When a predetermined timing occurs while the user is inputting chat text to the chat text input means,
The chat voice data transmission means transmits chat voice data generated by voice synthesis only for the chat text that has already been input to the online conference server, and
The speech synthesis means controls the speech synthesis speech tempo to be slowed down in consideration of the text chat that will be input subsequently.
The present invention is characterized in that the computer is made to function in such a manner .
According to the present invention, there is also provided a program for causing a computer installed in a terminal that exchanges voice data in real time via an online conference server to function, the program comprising:
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing;
A speech recognition means for converting conference speech data received from an online conference server into speech text;
a similarity calculation means for calculating a similarity between appearing word strings as a degree of agreement between the contents of the conversation between the spoken voice text and the chat text;
a filler voice data transmitting means for transmitting different filler voice data according to the similarity before the chat voice data transmitting means transmits the chat voice data;
to make the computer function
The filler audio data transmitting means includes:
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
The present invention is characterized in that the computer further functions as follows.

According to another embodiment of the program of the present invention,
The method further comprises: detecting a conversation gap during which the speech voice data of the conference received from the online conference server is not received for a predetermined period of time or longer;
It is also preferable that the chat voice data transmitting means causes the computer to function so as to transmit the chat voice data to the online conference server when a conversation gap is detected as the predetermined timing.

According to another embodiment of the program of the present invention,
The terminal further includes a microphone that can be turned on/off by a user's operation to collect the user's speech data;
It is also preferable to make the computer function so that the voice synthesis means and the chat voice data transmission means function when the microphone is turned off.

According to another embodiment of the program of the present invention,
The terminal is further equipped with a motion detection sensor,
It is also preferable that the computer further functions so that, when it is determined by the movement detection sensor that the user is moving, the voice synthesis means and the chat voice data transmission means function.

According to another embodiment of the program of the present invention,
A speech recognition means for converting conference speech data received from an online conference server into speech text;
It is also preferable that the computer further functions as a statement request determination means for determining whether or not the spoken voice text contains a predetermined keyword that requests a statement from the user of the terminal.

According to the present invention, there is provided a terminal for exchanging voice data in real time via an online conference server, comprising:
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing ;
When a predetermined timing occurs while the user is inputting chat text to the chat text input means,
The chat voice data transmission means transmits chat voice data generated by voice synthesis only for the chat text that has already been input to the online conference server, and
The speech synthesis means controls the speech synthesis speech tempo to be slowed down in consideration of the text chat that will be input subsequently.
It is characterized by:
According to the present invention, there is also provided a terminal for exchanging voice data in real time via an online conference server, comprising:
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing;
A speech recognition means for converting conference speech data received from an online conference server into speech text;
a similarity calculation means for calculating a similarity between appearing word strings as a degree of agreement between the contents of the conversation between the spoken voice text and the chat text;
a filler voice data transmitting means for transmitting different filler voice data according to the similarity before the chat voice data transmitting means transmits the chat voice data;
having
The filler audio data transmitting means includes:
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
It is characterized by:

According to the present invention, there is provided a chat voice data transmission method for terminals exchanging voice data in real time via an online conference server, comprising the steps of:
The terminal is
A first step of accepting input of chat text by a user operation;
a second step of synthesizing the chat text into chat voice data;
and a third step of transmitting chat voice data to an online conference server at a predetermined timing .
In the first step, when a predetermined timing occurs while the user is inputting chat text,
In the third step, chat voice data generated by voice synthesis only for the chat text already input is transmitted to the online conference server,
In the second step, the speech synthesis tempo is controlled to be slower, taking into account the text chat that will follow.
It is characterized by:
According to the present invention, there is also provided a chat voice data transmission method for terminals exchanging voice data in real time via an online conference server, comprising the steps of:
The terminal is
A first step of accepting input of chat text by a user operation;
a second step of synthesizing the chat text into chat voice data;
a third step of transmitting chat voice data to an online conference server at a predetermined timing;
A fourth step of converting the conference speech voice data received from the online conference server into speech voice text;
a fifth step of calculating a similarity between appearing word strings as a degree of agreement between the conversation contents of the spoken voice text and the chat text;
A sixth step of transmitting different filler voice data according to the similarity before transmitting the chat voice data for the third step;
Run
Regarding the sixth step,
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
It is characterized by:

The program, terminal, and method of the present invention can assist users who cannot speak during online conferences.

FIG. 1 is a schematic diagram of an online conference system according to the prior art. FIG. 2 is a screen diagram of the online conference system shown in FIG. 1 . FIG. 2 is a first functional configuration diagram of a terminal according to the present invention. FIG. 2 is a screen diagram of the online conference system according to the present invention. FIG. 2 is a second functional configuration diagram of the terminal according to the present invention. FIG. 3 is a third functional configuration diagram of the terminal according to the present invention. FIG. 4 is an explanatory diagram showing the transmission of filler audio data in the present invention. FIG. 4 is a fourth functional configuration diagram of the terminal according to the present invention. FIG. 2 is an explanatory diagram showing detection of a speech request in the present invention.

The following describes in detail the embodiments of the present invention with reference to the drawings.

FIG. 3 is a first functional configuration diagram of a terminal according to the present invention.
FIG. 4 is a diagram showing a screen of the online conference system according to the present invention.

As shown in FIG. 3, terminal 1 is operated by a user who will be a participating member of the online conference. An application for the online conference system is pre-installed on terminal 1. The application on terminal 1 functions as a user interface for the online conference, and also communicates with online conference server 2 to exchange voice data in real time with other members of the conference.

The terminal 1 has, as a user interface, a touch panel display 101, an audio output unit (speaker or audio output terminal) 102, and a microphone 103.
The touch panel display 101 displays images of the members participating in the online conference through an application, and also displays an input interface for chat text that can be input by the user (see FIG. 4, for example).
The audio output unit 102 outputs audio data uttered by the members participating in the online conference to the user as an audio signal by the application. For example, as shown in FIG. 1, in an environment where the user cannot speak, such as when the user is traveling on a train, the user can use earphones to listen to the discussion between the members of the online conference. Of course, when the user is in an environment where he or she can speak, the audio of the online conference can also be heard from a speaker.
The microphone 103 collects voice data uttered by the user who operates the terminal 1. The microphone 103 can be turned on/off by the user. For example, in an environment where the user cannot speak, the microphone 103 is turned off.

As shown in FIG. 3, the terminal 1 has a chat text input unit 11, a voice synthesis unit 12, a chat voice data transmission unit 13, a conversation gap detection unit 14, and a chat text transmission unit 15. These functional components are realized by executing a program that causes a computer installed in the terminal to function. The processing flow of these functional components can also be understood as a voice data transmission method of the terminal.

[Chat text input unit 11]
The chat text input unit 11 accepts input of chat text by a user operation. As shown in Fig. 4, the user can input chat text in the right column. The input chat text is output to the voice synthesis unit 12.

[Speech synthesis unit 12]
The voice synthesis unit 12 synthesizes chat text into chat voice data. This is, for example, an existing technology called Text to Speech, and converts text into real voice. In particular, it converts the quality of the voice into a voice that sounds natural, like a human speaking. At this time, it is preferable to generate a synthetic voice that is close to the characteristics of the user's voice (male/female, high/low pitch, etc.). The synthesized chat voice data is output to the chat voice data transmission unit 13.

[Chat voice data transmission unit 13]
The chat voice data sending unit 13 sends the chat voice data to the online conference server 2 at a predetermined timing. Here, the "predetermined timing" may be a suitable timing for the user to speak, using techniques such as those disclosed in Patent Documents 1 and 2 described above as conventional techniques.

According to the display screen of the online conference in FIG. 4, chat text entered by user @a is displayed in the right column, and a synthetic voice that sounds like user @a is spoken at a specified timing. The synthetic voice spoken during the online conference can be heard by other members as if it were spoken by user @a.

Note that, according to FIG. 3, it is preferable that the voice synthesis unit 12 and the chat voice data transmission unit 13 function only when the microphone is turned off. If the microphone is turned off, it can be determined that user @a is in an environment where he or she cannot speak, even though he or she is participating in an online conference.

[Conversation Gap Detection Unit 14]
The conversation gap detection unit 14 detects a "conversation gap" when the conference speech voice data received from the online conference server 2 has not been received for a predetermined period of time or more. That is, it detects a moment (a gap) during a conference when the discussion among multiple members stops for an instant. It determines that this is a suitable timing for the user to speak. It notifies the chat voice data transmission unit 13 of this timing as a predetermined timing.
Thereby, the chat voice data sending unit 13 sends the chat voice data to the online conference server 2 when a conversation gap is detected as the predetermined timing.

[Chat text transmission unit 15]
The chat text sending unit 15 sends the chat text to the online conference server 2. The chat text sending unit 15 may function only when the microphone is off, or may function normally whether the microphone is on or off.

As another embodiment of FIG. 3, the predetermined timing may occur while the user is inputting chat text. In such a case, the chat voice data sending unit 13 may send chat voice data that is voice-synthesized only for the chat text that has already been input to the online conference server 2. Whether chat text is being input can also be determined based on whether a character input interface is active. For example, when the chat text input screen is at the top, the character input interface may also be active.

4, for example, the following can be spoken during an online conference: "" is spoken voice data, and { } is input chat text.
User @c: "X company's equipment is cheap."
User @b: "But company Y's management functions are excellent."
User @a {Even though it's expensive, considering the maintenance}...
Conversation Gaps Detection
User @a: "Even though it's expensive, when you consider the maintenance..."
In this way, for example, during a gap in conversation (predetermined timing), the user can quickly speak out the opinion that he or she inputted in the chat text during the conference.

At this time, it is also preferable that the voice synthesis unit 12 controls the speech synthesis to slow down the speech tempo, taking into consideration the text chat that will be input subsequently.
User @a {Even though it's expensive, considering the maintenance,}
Conversation Gaps Detection
User @a: "It may be expensive, but considering the maintenance..."
User@a{select company y}
User @a: "Choose company Y..." *Slow speech tempo User @a {Should}
User @a: "Yes, you should."
In this way, for example, by slowing down the speech tempo of the voice synthesis, it is possible to assist in the slowness of the user's chat text. In other words, even if the pauses in speech are extended, the user can speak the opinions he is inputting in the chat text as much as possible in the conference. For example, the voice synthesis unit 12 temporarily buffers the input chat text and converts it into synthetic voice, but it may also be a unit that detects the free space of this buffer. The closer the buffer is to being empty, the slower the input of the user's chat text is, so voice synthesis is performed to slow down the speech tempo.

Figure 5 is a diagram showing the second functional configuration of a terminal according to the present invention.

5, compared to FIG. 3, the terminal 1 further includes a movement detection sensor 104. The movement detection sensor 104 may be an acceleration sensor or a positioning sensor such as a Global Positioning System (GPS). This allows the terminal 1 to determine that the user is in an environment where he or she cannot speak in a conference.
The voice synthesis unit 12 and the chat voice data transmission unit 13 function when the movement detection sensor 104 determines that the user is moving.

Figure 6 is a diagram showing the third functional configuration of a terminal according to the present invention.

As compared to FIG. 3, FIG. 6 further includes a voice recognition unit 16, a similarity calculation unit 17, and a filler voice data transmission unit 18.

[Speech recognition unit 16]
The speech recognition unit 16 converts the utterance voice data of the conference received from the online conference server 2 into utterance voice text. The converted utterance voice text is output to the similarity calculation unit 17.

[Similarity calculation unit 17]
The similarity calculation unit 17 calculates the degree of similarity between the conversation content of the speech text (text spoken by other members in a conference) and the chat text (text entered by a user who is in an environment where he or she cannot speak) by calculating ... Calculate the similarity of columns.
This is a common language processing technique that breaks down text into multiple words using morphological analysis, and then converts each text into a vector. If the similarity between the words is equal to or greater than a certain threshold, It is determined whether or not

For example, each unique word can be vectorized using Word2vec (registered trademark). "Word2vec" refers to a technology that compresses the dimension by expressing words as vectors in order to capture the meaning and grammar of the words. Alternatively, a bag of words may be used to express only the frequency of occurrence of each word contained in a sentence as a vector. Of course, it is preferable to compare the similarity between the entire text (groups of unique words) rather than the match rate between unique words.

It is not limited to Word2vec or Bag of Words, but may be converted into a feature vector obtained by analyzing the part of speech or meaning of each word. For spoken voice text, topic segmentation may be performed by, for example, TextTiling, and the segmented text may be vectorized.
S(a,b)=cosθ=(Va・Vb)/(|Va||Vb|)
S(a,b): Cosine similarity
Va: Text of speech by other members of the meeting
Vb: Chat text entered by the user. The cosine similarity S(a, b) is a value between 0 and 1, and the closer to 1 the similarity is, the higher it is.

[Filler audio data transmission unit 18]
A filler voice data transmitting unit transmits different filler voice data according to the degree of similarity before the chat voice data transmitting unit transmits the chat voice data.
Here, "filler" means something that fills in or fills something up, and in the field of language, it refers to words and phrases such as "um,""ah," and "hmm" that are inserted semi-unconsciously between gaps in a conversation. In the present invention, fillers are words that are spoken before speaking in a meeting.

The filler audio data transmitting unit 18 uses different fillers as follows depending on whether the similarity is equal to or greater than a predetermined threshold (true) or not (false).
True: The similarity is equal to or greater than a predetermined threshold, which means that the topic of the speech text of other members in the conference is the same as the chat text entered by the user. In this case, the first filler speech data is sent based on a filler word that is desired to be spoken along with the current topic. For example, "May I make a comment?", "Can I just say something?", etc.
False: If the similarity is not equal to or greater than the predetermined threshold, it means that the topic of the chat text entered by the user is behind the speech text of other members in the conference. In this case, a second filler voice data based on a filler word that indicates a desire to return to the previous topic is sent. For example, "May I make a comment about the previous matter?", "I'm going back," etc.

Of course, if the user has not finished inputting chat text even though filler audio data has been sent, it is preferable to prompt the user to complete input on the text input screen. This allows the user to speak their opinion in the conference as quickly as possible.

Figure 7 is an explanatory diagram showing the transmission of filler audio data in the present invention.

According to Fig. 7(a), filler voice data is transmitted, for example, as follows: "" is spoken voice data, and { } is input chat text.
User @c: "It's expensive, but considering the maintenance..."
User @b: "Company Y's management functions are excellent."
User @a {Even though it is expensive, considering the maintenance, we should choose company y.}
*The speech text of other members in the conference and the chat text of user @a are determined to have a high degree of similarity (above a predetermined threshold).
User @a (filler audio data)
"May I speak?"
User @a: "Even though it's expensive, if you consider the maintenance, you should choose company Y."

According to FIG. 7B, for example, filler audio data is transmitted as follows:
User @c: "It's expensive, but considering the maintenance..."
User @b: "Company Y's management functions are excellent."
User @c: "By the way, how's the situation with Company W going?"
User @b: "I've already contacted W company."
User @a {Even though it is expensive, considering the maintenance, we should choose company y.}
*The speech text of other members in the conference and the chat text of user @a are determined to have a low similarity (less than a predetermined threshold).
User @a (filler audio data)
"May I speak about what we discussed earlier?"
User @a: "Even though it's expensive, if you consider the maintenance, you should choose company Y."

Figure 8 is a fourth functional configuration diagram of a terminal according to the present invention.

As compared to FIG. 3, FIG. 8 further includes a voice recognition unit 16 and a speech request determination unit 19. The voice recognition unit 16 is the same as that described above.

[Speech request determination unit 19]
The speech request determination unit 19 determines whether or not the uttered voice text contains a predetermined keyword that requests the user of the terminal to make a speech. The predetermined keyword may be registered in advance for each user. .

Figure 9 is an explanatory diagram showing the detection of a request to speak in the present invention.

9, for example, a user's name [a] is registered in the speech request determination unit 19. For example, assume that the following spoken voice text is detected: "" is the spoken voice data, and { } is the input chat text.
User @c: "X company's equipment is cheap."
User @b: "But company Y's management functions are excellent."
User @a {Even though it is expensive, considering the maintenance, we should choose company y.}
User @c: "What do you think, A?"
At this time, the speech request determination unit 19 detects [a] and determines that a request has been made to the user @a to speak in the conference. Then, the speech request determination unit 19 instructs the chat voice data transmission unit 13 to transmit chat voice data.
Then, chat voice data in the form of synthetic voice like the following will be spoken in the conference.
User @c: "Even though it's expensive, if you consider the maintenance, you should choose company y."

As described above in detail, the program, terminal, and method of the present invention can assist users who are unable to speak during an online conference. In particular, the opinions of users who are in an environment where they cannot speak can be communicated to all members at the most appropriate time possible.

As a result, for example, it will be possible to "provide online meetings regardless of the user's location," which will contribute to Goal 8 of the United Nations' Sustainable Development Goals (SDGs), which is to "promote inclusive and sustainable economic growth, employment and decent work for all."

With respect to the various embodiments of the present invention described above, various changes, modifications, and omissions within the scope of the technical ideas and viewpoints of the present invention can be easily made by a person skilled in the art. The above description is merely an example and is not intended to be restrictive in any way. The present invention is limited only by the scope of the claims and their equivalents.

REFERENCE SIGNS LIST 1 Terminal 101 Touch panel display 102 Voice output unit 103 Microphone 104 Movement detection sensor 11 Chat text input unit 12 Voice synthesis unit 13 Chat voice data transmission unit 14 Conversation gap detection unit 15 Chat text transmission unit 16 Voice recognition unit 17 Similarity calculation unit 18 Filler voice data transmission unit 19 Speech request determination unit 2 Online conference server

Claims (10)

A program for causing a computer installed in a terminal that exchanges voice data in real time via an online conference server to function,
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
causing the computer to function as chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing ;
When a predetermined timing occurs while the user is inputting chat text to the chat text input means,
The chat voice data transmission means transmits chat voice data generated by voice synthesis only for the chat text that has already been input to the online conference server, and
The speech synthesis means controls the speech synthesis speech tempo to be slowed down in consideration of the text chat that will be input subsequently.
A program that causes a computer to function in such a manner . A program for causing a computer installed in a terminal that exchanges voice data in real time via an online conference server to function,
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing;
A speech recognition means for converting conference speech data received from an online conference server into speech text;
a similarity calculation means for calculating a similarity between appearing word strings as a degree of agreement between the contents of the conversation between the spoken voice text and the chat text;
a filler voice data transmitting means for transmitting different filler voice data according to the similarity before the chat voice data transmitting means transmits the chat voice data;
to make the computer function
The filler audio data transmitting means includes:
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
A program for causing a computer to further function as described above. The method further comprises: detecting a conversation gap during which the speech voice data of the conference received from the online conference server is not received for a predetermined period of time or longer;
3. The program according to claim 1, wherein the chat voice data transmission means causes the computer to function so as to transmit chat voice data to the online conference server when a conversation gap is detected as the predetermined timing. The terminal further includes a microphone that can be turned on/off by a user's operation to collect the user's speech data;
4. The program according to claim 1, further comprising: causing a computer to function so that a voice synthesis unit and a chat voice data transmission unit function when a microphone is turned off. The terminal is further equipped with a motion detection sensor,
4. The program according to claim 1, further causing the computer to function so that a voice synthesis unit and a chat voice data transmission unit function when the movement detection sensor determines that the user is moving. A speech recognition means for converting conference speech data received from an online conference server into speech text;
The program according to any one of claims 1 to 5 , further causing the computer to function as a speech request determination means for determining whether or not the spoken voice text contains a predetermined keyword that requests a statement from the user of the terminal. A terminal for exchanging voice data in real time via an online conference server,
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing ;
When a predetermined timing occurs while the user is inputting chat text to the chat text input means,
The chat voice data transmission means transmits chat voice data generated by voice synthesis only for the chat text that has already been input to the online conference server, and
The speech synthesis means controls the speech synthesis speech tempo to be slowed down in consideration of the text chat that will be input subsequently.
A terminal characterized by: A terminal for exchanging voice data in real time via an online conference server,
a chat text input means for accepting input of a chat text by a user's operation;
A voice synthesis means for synthesizing chat text into chat voice data;
a chat voice data transmission means for transmitting chat voice data to an online conference server at a predetermined timing;
A speech recognition means for converting conference speech data received from an online conference server into speech text;
a similarity calculation means for calculating a similarity between appearing word strings as a degree of agreement between the contents of the conversation between the spoken voice text and the chat text;
a filler voice data transmitting means for transmitting different filler voice data according to the similarity before the chat voice data transmitting means transmits the chat voice data;
having
The filler audio data transmitting means includes:
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
A terminal characterized by: A chat voice data transmission method for terminals exchanging voice data in real time via an online conference server, comprising:
The terminal is
A first step of accepting input of chat text by a user operation;
a second step of synthesizing the chat text into chat voice data;
and a third step of transmitting chat voice data to an online conference server at a predetermined timing .
In the first step, when a predetermined timing occurs while the user is inputting chat text,
In the third step, chat voice data generated by voice synthesis only for the chat text already input is transmitted to the online conference server,
In the second step, the speech synthesis tempo is controlled to be slower, taking into account the text chat that will follow.
A chat voice data transmission method for a terminal, comprising: A chat voice data transmission method for terminals exchanging voice data in real time via an online conference server, comprising:
The terminal is
A first step of accepting input of chat text by a user operation;
a second step of synthesizing the chat text into chat voice data;
a third step of transmitting chat voice data to the online conference server at a predetermined timing ;
A fourth step of converting the conference speech voice data received from the online conference server into speech voice text;
a fifth step of calculating a similarity between appearing word strings as a degree of agreement between the conversation contents of the spoken voice text and the chat text;
A sixth step of transmitting different filler voice data according to the similarity before transmitting the chat voice data for the third step;
Run
Regarding the sixth step,
When the similarity is determined to be equal to or greater than a predetermined threshold, a first filler voice data based on a filler word that is desired to be spoken in accordance with the current topic is transmitted;
When the answer is false, a second filler voice data based on a filler word indicating a desire to return to the previous topic is transmitted.
A chat voice data transmission method for a terminal, comprising:

Images