 |
Description  |
|
|
BACKGROUND OF THE
INVENTION
1. Field of the Invention
The present invention relates to an audio interactive system, in which a plurality of users may interact with one another through a speech, in an on-line chat system utilizing a computer and a computer network.
2. Description of the Related Art
Popularization of the computer network makes it easy to do works, to play games, and to exchange life information through conversations of a plurality of users on a network.
Thus, there has been required more simple input means in place of keyboards so that even ordinary users can attend such a network conversation system (a chat system).
According to the conventional chat system, a sentence is inputted through operation of a keyboard, and a conversation is effected by transmitting the entered sentence.
However, it would be very difficult for ordinary users to operate the keyboard as to Japanese syllabary-to-Chinese character transformation and the like. This causes the input speed of the keyboard operation to slow down, and thus involves such
a problem that the operator cannot follow a flow of the conversation. In order to solve this problem, there has been developed a system in which a conversation is effected through a speech.
FIG. 61 is a schematic construction view of the related art of audio interactive system. Here, for the purpose of a simplification, there are shown only two terminals A and B, and it is assumed that a conversation is made between the terminals A
and B.
Each of the terminals A and B comprises audio input means 11A and 11B each for converting voice into an aural signal, such as a microphone and the like, and audio output means 12A and 12B each for converting an aural signal into voice, such as a
speaker and the like. In such an audio interactive system, the terminals A and B are connected with one another via a communication line 1 under control of a router 2 so that a conversation is made between the terminals A and B by transmitting and
receiving the aural signals or the sound signals.
However, the above-mentioned audio interactive systems are short of a bandwidth of a communication channel, and thus there simply exist systems which permit conversation between 2-4 subscribers.
Consequently, in order to allow a plurality of users to have a conversation through a network, it will be a subject that how the aural data, which are increased as the users are increased, are sliced.
SUMMARY OF THE INVENTION
In view of the foregoing, it is therefore an object of the present invention to provide an audio interactive system which is preferable for suitably controlling a communication amount among the users.
To attain the above-mentioned object, according to the present invention, there is provided an audio interactive system comprising: a plurality of terminals each having an audio input means for converting voices into aural signals and an audio
output means for converting aural signals into voices, said plurality of terminals being connected to a communication line; and a server connected with said plurality of terminals via said communication line adapted to perform collection and distribution
of the aural signals, in which a conversation is made through voices between said plurality of terminals, wherein said server comprises a buffer means for temporarily storing the aural signals transmitted from the terminals, and a scheduler means for
controlling the distribution of the aural signals stored in said buffer means.
Here, it is acceptable that said server controls a distribution sequence of the aural signals.
FIG. 1 is a principle view showing this aspect of the present invention.
In FIG. 1 and the following figures except some figures, only two terminals are typically illustrated.
As shown in FIG. 1, speeches transmitted from terminals A and B are stored in buffer memories 21A and 21B of a server 20 under control of a control unit 23, respectively. The speeches are ordered suitably by a scheduler 22, and sequentially
distributed to the terminals A and B. For the purpose of better understanding, the buffer memories are illustrated in the same number as those of the terminals A and B.
According to this aspect, it is possible to avoid such a problem that when speeches take place simultaneously from a plurality of terminals A and B, parallel incoming of these speeches to the terminals A and B causes a bandwidth to exceed that of
a line or circuit, so that the speeches unnaturally break. Further, it is possible to save trouble of processing for parallel incoming of a plurality of speeches to the terminals A and B and processing such that a plurality of speeches are mixed in the
terminals A and B. Incidentally, the terminals A and B have control units 13A and 13B, respectively.
In the audio interactive system according to the present invention, it is preferable that said server has means for managing a number of simultaneous speakers who simultaneously transmit speeches, and inhibits a simultaneous transmission of the
aural signals from the terminal, a number of said aural signals exceeding a predetermined number.
FIG. 2 is a principle view showing this aspect.
As shown in FIG. 2, in a case where speeches are made from terminals A and B, each of speech requirement means 14A and 14B informs a server 20 of a speech requirement. In the server 20, a number of persons who can speak simultaneously is set up
beforehand to a simultaneous speaker number management means 24 for managing a number of simultaneous speakers, taking account of a bandwidth of line 1. For example, if it is permitted up to three persons, then number 3 is set up to the simultaneous
speaker number management means 24. In this condition, when the server 20 receives speech requirements from the terminals, the server 20 confirms the numerical value stored in the simultaneous speaker number management means 24. If the numerical value
is 1 or more, the server 20 gives a notice of allowance for the terminals A and B, and the numerical value is reduced by 1. If the numerical value is already 0, the server 20 waits until a notice of speech termination arrives from some terminal.
In the terminal ends, upon receipt of a notice of allowance, it informs the users through speech allowance display means 15A and 15B that they are allowed to make speeches. After the users confirm the display, they start the speeches, and after
termination of their speeches, they inform the server 20 of termination of their speeches through speech termination informing means 16A and 16B.
In the server end, upon receipt of a notice of termination of the speeches, the numerical value stored in the simultaneous speaker number management means 24 is increased by 1.
In an audio interactive system according to the present invention, it is preferable that the server or the terminal has a silence detection means for detecting a silence portion of the aural signal, and said scheduler controls a distribution
sequence of the aural signals in units of speeches each separated with the silence portion of the aural signal.
FIG. 3 is a principle view showing this aspect.
As shown in FIG. 3, speeches transmitted from terminals A and B are stored in buffer memories 21A and 21B of a server 20 under control of a control unit 23, respectively. PCM data stored in the buffer memories 21A and 21B are searched by a
silence detection means 25 to measure a sound volume. If a period of time during which the measured sound volume is smaller than a predetermined value is longer than a predetermined time, the speech is partitioned on that portion. The speeches thus
partitioned are ordered suitably by a scheduler 22, and sequentially distributed to the terminals A and B.
According to this aspect, in addition to the fundamental advantages of the present invention, it is possible to detect pauses of a speech and switch the speech to another speech at the detected pause, thereby avoiding useless occupation of the
system in the silence portion.
In an audio interactive system according to the present invention, it is preferable that said scheduler controls a distribution sequence of the aural signals in such a manner that the server distributes the aural signals in the order in which the
aural signals arrive at said server.
FIG. 4 is a principle view showing this aspect.
As shown in FIG. 4, speeches transmitted from terminals A and B are stored in buffer memories 21A and 21B of a server 20 under control of a control unit 23, respectively. And in addition, receipt starting times of speeches are stored in the
buffer memories 21A and 21B. In the scheduler 22, if there is speech now in transmission, the transmission is continued. On the other hand, if the transmission is over, the earlier one of receipt starting times of all the buffer memories 21A and 21B is
transmitted to all the terminals up to a pause of the speech.
According to this aspect, in addition to the fundamental advantages of the present invention, there is an advantage such that conversations along the time base are possible.
In an audio interactive system according to the present invention, it is preferable that said server has means for transmitting to the terminal a command indicating transmission of an internal time of the terminal; each of said terminals has
means responsive to the command for transmitting one's own internal time to said server, and means for transmitting to said server the aural signal together with one's own internal time at time of transmission; and said scheduler controls a distribution
sequence of the aural signals in such a manner that the server distributes the aural signals in the order in which the aural signals are generated from said terminals.
FIG. 5 is a principle view showing this aspect.
As shown in FIG. 5, at the time point when terminals A and B are connected to a server 20, an internal time correction value computing unit 27 within the server 20 is activated. First, the server 20 instructs the terminals A and B that local
times of the terminals A and B are read through internal clocks 17A and 17B and the local times thus read are transmitted to the server 20. Now it is assumed that a time of the instruction is expressed by Ts. The server 20 waits so that the terminals A
and B responsive to the instruction report the local times of the terminals A and B. When the server 20 received the reports from the terminals A and B, this time is expressed by Te. Assuming that the reported local times of the terminals A and B are
expressed by T1, difference between an internal clock 26 within the server 20 and the internal clock 17A or 17B of the terminals A or B is expressed by an internal time correction value Td=Ts+(Te-Ts)/2-T1. This value is evaluated on all of the terminals
A and B, and each of the obtained values is stored in the associated one of internal time correction value memories of buffers 21A and 21B.
In each of the terminals A and B, time Tx at which speech is made is counted by the internal clock 17A or 17B, and the speech is transmitted to the server 20 together with the time Tx. This feature makes it possible for the server 20 to correct
the time, at which the speech transmitted from each of the terminals A and B was actually made, to a time of the server internal clock by Tx-Td.
When the speeches generated from the terminals A and B are stored in the buffers 21A and 21B, the internal times transmitted from the terminals A and B are corrected to the server internal times, using internal time correction values, and then
stored in the associated corrected receipt starting time memories in the server 20, respectively. Thus, the scheduler 22 can transmit sequentially the speeches to all the terminals A and B in accordance with the order in which the speeches were actually
made.
According to this aspect, in addition to the fundamental advantages of the present invention, there is an advantage such that speeches are permitted sequentially in accordance with the order in which the speeches actually take place. Thus, it is
possible to provide fairly a service which is to be made in earlier order, for example, a quiz.
In an audio interactive system according to the present invention, it is preferable that said server has a mixer for mixing the aural signals.
FIG. 6 is a principle view showing this aspect.
As shown in FIG. 6, speeches transmitted from terminals A and B are stored in buffer memories 21A and 21B of a server 20, respectively. And these speeches are mixed by a mixer 29 into a single speech and temporarily stored in an output buffer
area 28, and thereafter it is distributed to all the terminals A and B.
According to this aspect, it is possible to avoid such a problem that when speeches take place simultaneously from a plurality of terminals A and B, parallel incoming of these speeches to the terminals A and B causes a bandwidth to exceed that of
a line or circuit, so that the speeches unnaturally break. Further, it is possible to save trouble of processing for parallel incoming of a plurality of speeches to the terminals A and B and processing such that a plurality of speeches are mixed in the
terminals A and B.
In an audio interactive system according to the present invention, it is preferable that said scheduler controls a distribution time of the aural signals on the basis of time at which the aural signals arrive at said server.
FIG. 7 is a principle view showing this aspect.
As shown in FIG. 7, speeches transmitted from terminals A and B are stored in buffer memories 21A and 21B of a server 20, respectively. And in addition, receipt starting times of speeches are stored in the buffer memories 21A and 21B. A mixer
29 arranges the respective speeches in time sequence referring to the receipt starting times, and mixes the same time of speeches and temporarily stores the mixed speech in an output buffer area 28. The scheduler 22 the speeches, which have been
subjected to mixing process in which the same time of speeches are mixed into a single speech, are distributed in order of time.
According to this aspect, in addition to the fundamental advantages of the present invention, it is possible to mix the speeches upon arranging the same in order of time.
In an audio interactive system according to the present invention, it is preferable that said said server has means for transmitting to the terminal a command indicating transmission of an internal time of the terminal; each of said terminals has
means responsive to the command for transmitting one's own internal time to said server, and means for transmitting to said server the aural signal together with one's own internal time at time of transmission; and said scheduler controls a distribution
time of the aural signals on the basis of time at which the aural signals are generated from said terminals.
FIG. 8 is a principle view showing this aspect.
As shown in FIG. 8, at the time point when terminals A and B are connected to a server 20, first, the server 20 instructs the terminals A and B that local times of the terminals A and B are read through internal clocks 17A and 17B and the local
times thus read are transmitted to the server 20. Now it is assumed that a time of the instruction is expressed by Ts. The server 20 waits that the terminals A and B responsive to the instruction report the local times of the terminals A and B. When
the server 20 received the reports from the terminals A and B, this time is expressed by Te. Assuming that the reported local times of the terminals A and B are expressed by T1, difference between an internal clock 26 within the server 20 and the
internal clock of the terminals A or B is expressed by an internal time correction value Td=Ts+(Te-Ts)/2-T1. This initial setting processing is first effected on all of the terminals A and B.
In each of the terminals A and B, time Tx at which speech is made is counted by the internal clock 17A or 17B, and the speech is transmitted to the server 20 together with the time Tx. This feature makes it possible for the server 20 to correct
the time, at which the speech transmitted from each of the terminals A and B was actually made, to a time of the server internal clock by Tx-Td.
When the speeches generated from the terminals A and B are stored in the buffers 21A and 21B, the internal times transmitted from the terminals A and B are corrected to the server internal times, using internal time correction values, and then
stored in the associated corrected receipt starting time memories in the server 20, respectively. Thus, a mixer 29 can mix the speeches in accordance with times at which the speeches were actually made. The scheduler 22 controls a distributing time in
accordance with the times at which the speeches were actually made.
According to this aspect, in addition to the advantages discussed on the aspect explained referring to FIG. 6, there is an advantage such that speeches are permitted sequentially in accordance with the order in which the speeches actually take
place. Thus, it is possible to provide fairly a service which is to be made in earlier order, for example, a quiz.
In an audio interactive system according to the present invention, it is preferable that each of said terminals has a logical personal image, and said scheduler controls a distribution of the aural signals on the basis of at least one of logical
positional relation, distance, surroundings, eyes direction, sight and posture of the logical personal image within space of the logical personal image.
FIG. 9 is a principle view showing this aspect.
As shown in FIG. 9, terminals A and B comprises avatar management information memories 17A and 17B each for storing avatar management information to manage position, eyes direction, sight and the like of a logical personal image (a so-called
avatar) representative of a user (speaker) of the terminals A or B, and world management information memories 18A and 18B for storing world management information to manage a logical world held in common by the avatars of the terminals A and B,
respectively. The users of the terminals A and B may control the position of the avatar representative of oneself, one's eyes direction and the like.
Each of the terminals A and B transmits the speech and in addition a moving direction of one's own avatar, a moving direction of one's eyes and the like. On the other hand, the server 20 stores the received speeches in the buffer memories 21A
and 21B, and successively update the contents of avatar management information memories for storing avatar management information to manage logical position, eyes direction, sight and the like of the avatar. The server 20 comprises a world management
information memory 30 which is similar to those in the terminals A and B, and a maximum distance memory 31 for storing a maximum distance indicating a maximum distance between avatar-to-avatar up to which a conversation is permissible. In the scheduler
22, when the speeches are sequentially transmitted to the terminals A and B, there is performed such a scheduling that a speech, which is issued from an avatar who cannot be seen (out of sight) from a certain speaker (avatar), is not transmitted,
otherwise it is transmitted with decreasing priority.
According to this aspect, in addition to the fundamental advantages of the present invention, there is an advantage that such an inconvenience that an avatar is addressed from a speaker who cannot been seen is avoided. Further, this feature
makes it possible to provide such a control that a speech issued from a speaker who is all too far apart cannot be heard. Furthermore, the limitation of distribution of speeches unnecessary to be heard makes it possible to reduce an unnecessary
communication amount.
In an audio interactive system according to the present invention, it is preferable that each of said terminals has a logical personal image (avatar), and has an effector for making an alteration to the aural signal on the basis of at least one
of logical positional relation, distance, surroundings, eyes direction, sight and posture of the logical personal image within space of the logical personal image.
FIG. 10 is a principle view showing this aspect.
As shown in FIG. 10, each of the terminals A and B transmits the speech and in addition a moving direction of a speaker (avatar), a moving direction of one's eyes and the like. On the other hand, the server 20 stores the received speeches in the
buffer memories 21A and 21B, and successively update the contents of avatar management information memories for storing avatar management information to manage logical position, one's eyes direction, sight and the like of the avatar. With respect to the
speeches stored in the buffer memories, when such speeches are distributed to the terminals, there is provided such a control that a speech issued from a speaker who is logically all too far apart is not distributed, and in case of other than that, the
speeches stored in the buffer memories are processed by an effector 32 in such a manner that the sound volume is smaller as a logical distance between the speaker and the listener is extended, taking into account of a logical positional relation between
the speaker and the listener and their eyes directions. Further, a sound volume ratio of the left and the right, a phase and a tone quality(characteristic of a filter) are varied in accordance with a direction. In a case where there is an obstacle
between the speaker and the listener, an echo is applied or a tone quality (characteristic of a filter) is varied, in order to express that sound is propagated indirectly. The sound signal or aural signal thus modified is temporarily stored in output
buffer areas L and R in the form of a stereo sound signal.
The thus processed speeches from the respective terminals are subjected to a scheduling process and are sequentially distributed to the respective terminals. Alternatively, the speeches involved in the same time are subjected to a mixing process
and are distributed to the respective terminals.
According to this aspect, it is possible to emphasize a realistic image such that the user may hear sound with the sound volume corresponding to the distance of the speaker and hear like that the sound emanates from a direction corresponding to
the direction in which the speaker exists.
In an audio interactive system according to the present invention, it is preferable that said server has one or more group management tables registering terminals, which belong to an interaction group in which a conversation is made, of said
plurality of terminals, and said scheduler controls a distribution of the aural signals in such a manner that received aural signals are distributed to terminals which belong to the same interaction group as the terminal which generated the received
aural signals.
FIG. 11 is a principle view showing this aspect.
As shown in FIG. 11, there is provided a group management table 33 comprising a receiving group table on each receiving group and a transmitting group table on each transmitting group. The receiving group table on each receiving group stores
user IDs of all users who belong to the group. The transmitting group table on each transmitting group stores group IDs of all users who belong to the group. Speeches transmitted from the terminals A and B are stored in buffer memories 21A and 21B,
respectively, and then ordered by a scheduler 22 and sequentially distributed to all the terminals which belong to the same group. Alternatively, as aforementioned, it is acceptable that the speeches involved in the same time are subjected to a mixing
process for each of the terminals which belong to the same group, and are distributed to the respective terminals.
In an audio interactive system according to the present invention, it is preferable that each of said terminals has a transmitting mode selection means for selecting anyone of a plurality of transmitting modes for designating a receiving terminal
as an accepting station to receive an aural signal originated.
FIG. 12 is a principle view showing this aspect.
As shown in FIG. 12, an object of speech is selected beforehand by transmitting mode selection means 19A, 19B from among the same group, the multiple address to a plurality of groups, and the specified user (or an user group).
If an object of speech is the same group only, it implies an ordinary conversation mode. In this case, there is performed the processing explained referring to FIG. 11.
If the multiple address to a plurality of groups is instructed to the server 20, the receiving group ID, which is desired to be an object of the multiple address, is added to the associated user's transmitting group table 35 of a group management
table 33. With respect to speeches transmitted from the terminals A and B, since it is possible to determine the receiving group ID through referring to the associated user's transmitting group table 35, the speeches are distributed to all the belonging
users of each receiving group.
If a speech to the specified user (or an user group) is instructed to the server 20, the designated user ID (an ID of user group) is added to the associated user's transmitting group table 35 of the group management table 33. With respect to
speeches transmitted from the terminals A and B, since it is possible to determine the receiving group ID and the user ID through referring to the associated user's transmitting group table 35, the speeches are distributed to all the belonging users of
each receiving group, and all the users (or user group).
In an audio interactive system according to the present invention, it is preferable that each of said terminals has a listening mode selection means for selecting anyone of a plurality of listening modes for designating a transmitting terminal as
a transmitting source.
FIG. 13 is a principle view showing this aspect.
As shown in FIG. 13, an object of listening is selected beforehand by the listening mode selection means 101B from among the same group, the simultaneous listening of a plurality of groups, and the specified user.
If an object of listening is the same group only, it implies an ordinary conversation mode. In this case, there is performed the processing explained referring to FIG. 11.
If the simultaneous listening of a plurality of groups is instructed to the server 20, one's user ID is added to an entry of the receiving group table 34 of a group, which is desired to be an object of the listening, of a group management table
33. With respect to speeches transmitted from the terminals A and B, the receiving group ID is determined through referring to the associated user's transmitting group table 35. Thus, the speeches are distributed to all the belonging users registered
on the receiving group table 34. As a result, the speeches may be distributed also to the user ID added to the receiving group table 34.
If a listening to the specified user (or an user group) is instructed to the server 20, one's user ID is added to the transmitting group table 30 of the designated user (or an user group). With respect to speeches transmitted from the terminals
A and B, the receiving group ID and the user ID are determined through referring to the associated user's transmitting group table 35. Thus, the speeches are distributed to all the belonging users registered on the receiving group table 34 and the
designated user. As a result, the speeches may be distributed also to the user ID added to the transmitting group table 35.
In an audio interactive system according to the present invention, it is preferable that each of said terminals has an interruption inhibit informing means for informing said server of an interruption inhibit to designate a terminal, on which a
reception of aural signals addressed and transmitted to oneself is rejected, of terminals as originating sources of the aural signals; and said server stops, when a transmitting destination of received aural signals is the terminal which informed of
interruption inhibit through said interruption inhibit informing means and an originating source of the received aural signals is the terminal designated by interruption inhibit information through said interruption inhibit informing means, a transfer of
the aural signals to the terminal as the transmitting destination.
FIG. 14 is a principle view showing this aspect.
As shown in FIG. 14, the respective terminals inform a server 20 of interruption inhibit so as to inhibit the server 20 from originating speeches from other than groups designated by users, and users.
In the server 20, upon receipt of a notice of the interruption inhibit, if it is involved in the interruption inhibit to the receiving group, an interruption inhibit flag of the associated group of the receiving group table 34 is turned on. On
the other hand, if it is involved in the interruption inhibit to one's own terminal, an interruption inhibit flag of the associated terminal ID of the transmitting group table 35. When a designation of a speech mode is issued from a certain user, the
interruption inhibit flag of the receiving group table 34 and the interruption inhibit flag of the associated terminal ID of the transmitting group table 35 are confirmed, and if their flags are turned on, the registration onto the transmitting group
table 35 is rejected.
In an audio interactive system according to the present invention, it is preferable that a password is assigned to said interaction group; each of said terminals has a password input means for inputting the password to transmit the same to said
server, and said server inspects the passwords transmitted from the terminals and causes the terminal transmitting a proper password to join an associated interaction group.
FIG. 15 is a principle view showing this aspect.
As shown in FIG. 15, each of the terminals is provided with a password input means 102A, 102B. A server 20 is provided with a security check means 37, a privilege level table 36 for each user, a receiving group table 34 and a transmitting group
table 35. Each of entries of the receiving group table 34 and the originating group table 35 is provided with a privilege level memory and a password memory. A privilege level is an indicator indicating as to what level of group the respective users
are allowed to attend in conversation. With respect to a password, only the users who know the password are allowed to attend the conversation.
Here, it is assumed that in order to listen the speech of another group to which a user does not belong, when the receiving group table 34 of another user, or the transmitting group table 35 of another user is rewritten in accordance with the
request from the user, the security check means 37 checks the privilege level and the password of the user of concern. And only in a case where it is proper request, a rewrite is performed.
In an audio interactive system according to the present invention, it is preferable that each of said terminals has a display means for displaying a speaker and a listener.
FIG. 16 is a principle view showing this aspect.
As shown in FIG. 16, each of the terminals is provided with a display means 105A, 105B, a coordinates input means 104A, 105B for designating a user of interest in speaking from among a plurality of users (avatars) displayed on the display means
105A, 105B, and a drawing means for performing drawing on the display means 105A, 105B. It is identified on the display means 105A, 105B that who speaks to whom.
Thus, if there are added means for indicating a party who is spoken to and means for identifying as to who speaks to whom, it is permitted that a speech is transmitted together with a user ID of a party to the server 20 through speaking while
indicating a party of interest in speaking when a speech takes place, and the server 20 distributes to the respective terminals data together with the speaker's user ID.
In this manner, each of the terminals reads the speaker's user ID and the user ID of a party so as to vary the illustration of the speaker (avatar) and the illustration of the party.
In an audio interactive system according to the present invention, it is preferable that said server has an effector for changing tone quality of the aural signals.
FIG. 17 is a principle view showing this aspect.
As shown in FIG. 17, if the user desires it, a speech is transmitted together with a silent flag to the server 20 by a mode setting means 106A, 106B for setting a mode as to whether or not a tone quality is changed. The server 20 transmits, when
receives a speech to which a silent flag is appended, the speech to each of the terminals in such a manner that the speech is changed in sound by passing through an effector 38 so that a speaker cannot be identified.
In an audio interactive system according to the present invention, it is preferable that said server has a sound effect buffer for storing a sound effect signal representative of a predetermined sound effect, and said scheduler means controls a
distribution of the aural signals transmitted and also controls a distribution of the aural signals stored in said sound effect buffer.
FIG. 18 is a principle view showing this aspect.
As shown in FIG. 18, a server 20 is provided with a sound effect buffer 39 for storing a sound effect, a sound effect management information memory 40 for storing information to manage a sound effect and an iterative interval memory 41 for
storing an iterative interval of the sound effect.
Here, a scheduling process or a mixing process is performed in such a manner that when PCM data of the sound effect other than a user's speech, for example, a store broadcasting and a station announce, its logical position and an iterative
interval are registered onto the server 20, the server 20 repeatedly distributes these sound effects at regular intervals. Thus, the sound effect is also distributed to the respective terminals, in a similar fashion to that of the speech, in the
intervals between the speeches.
In an audio interactive system according to the present invention, it is preferable that said server or said terminal has audio recognition means for converting the aural signals into text data.
FIG. 19 is a principle view showing this aspect.
As shown in FIG. 19, when speeches transmitted from a terminal are distributed to the respective terminals, in a case where the speeches are distributed to a terminal having no audio output means, PCM data of the speech stored in the buffer is
converted into a text using a speech recognition means 42 incorporated into the server 20. Thus, in the terminal having no audio output means, it is possible to indicate the speeches with the text. It is acceptable that the terminal is provided with a
speech recognition means 107.
FIG. 20 is a principle view showing another aspect.
As shown in FIG. 20, the speech received in form of a text from a terminal having no audio input means is converted into PCM data using a voice synthesizer means 43 incorporated into the server 20, and for a terminal having audio output means the
data thus converted is distributed in form of a voice. On the other hand, for a terminal having no audio output means the speech received in form of a text from | | |
