Communication system for building speech database for speech synthesis, relay device therefor, and relay method therefor

A communication system comprising:

• a relay device connected to a communication network;
• at least two communication terminals connected to the communication network via the relay device, each communication terminal transmitting to, and receiving speech data from, another communication terminal via the relay device, wherein the at least two communication terminals includes a first communication terminal and a second communication terminal;
• and
• a media processing device connected to the relay device,
• the relay device comprising:
• a transmitter-receiver that receives first speech data originating from the first communication terminal and that transmits the received first speech data to the second communication terminal;
• a data duplicator that duplicates speech data;
• and
• a communication controller that causes the data duplicator to duplicate the first speech data received from the first communication terminal via the transmitter-receiver and that causes the transmitter-receiver to transmit the duplicated speech data to the media processing device, and
• the media processing device comprising:
• a receiver that receives, from the relay device, the duplicated speech data of the first communication terminal;
• a speech data processor that stores speech data received by the receiver in a speech data storage device, wherein the speech data processor further comprises a determiner that determines whether a piece of speech data received by the receiver corresponds to any piece of the stored speech data, and a noise measurer that measures the amount of noise contained in the received piece of speech data and the amount of noise contained in a corresponding piece of stored speech data, and wherein the speech data processor is configured to overwrite the corresponding piece of stored speech data with the received piece of speech data in a case in which an amount of noise of the received piece of speech data is less than that of the corresponding piece of stored speech data;
• a speech synthesis database generator that generates a speech synthesis database for the first communication terminal based on the speech data stored in the speech data storage device;
• a speech synthesis database storage device that stores the speech synthesis database generated by the speech synthesis database generator;
• and
• a speech synthesizer that executes speech synthesis based on the speech synthesis database in a case in which a request for speech synthesis is received from the first communication terminal.

A relay device for use in a communication system with a first communication terminal and a second communication terminal connected to the communication system via the relay device, the relay device comprising:

• a transmitter-receiver that receives speech data from the first communication terminal and transmits the received speech data to the second communication terminal;
• a data duplicator that duplicates speech data;
• and
• a communication controller that causes the data duplicator to duplicate the speech data received from the first communication terminal via the transmitter-receiver and that causes the transmitter-receiver to transmit the duplicated speech data together with identification information identifying one of the first or the second communication terminals as an originating communication terminal to a media processing device, the media processing device for storage of the duplicated speech data, and generation of a speech synthesis database, the media processing device including a speech data processor configured to measure a first acoustic quality associated with the duplicated speech data, and measure a second acoustic quality associated with a piece of previously stored speech data to determine whether the piece of previously stored speech data corresponds to the duplicated speech data.

A relay method for use at a relay device in a communication system with a first communication terminal and a second communication terminal connected to the communication system via the relay device, the method comprising:

• receiving speech data from the first communication terminal and transmitting the received speech data to the second communication terminal;
• duplicating the speech data received in the receiving step;
• and
• transmitting the duplicated speech data together with identification information identifying one of the first or the second communication terminals as an originating communication terminal to a media processing device, the media processing device storing the duplicated speech data and generating a speech synthesis database, the media processing device including a speech data processor configured to measure a first acoustic quality associated with the duplicated speech data, and measure a second acoustic quality associated with a piece of previously stored speech data to determine whether the piece of previously stored speech data corresponds to the duplicated speech data.

A media processing device comprising:

• a receiver configured to receive speech data from a relay device;
• a speech data processor configured to store the received speech data in a speech data storage device, wherein the speech data processor is further configured to determine whether a piece of the received speech data corresponds to a piece of previously stored speech data included in the speech data storage device by measurement of a first acoustic quality associated with the piece of received speech data, and measurement of a second acoustic quality associated with the piece of previously stored speech data;
• a speech synthesis database generator that generates a speech synthesis database based on speech data stored in the speech data storage device;
• and
• a speech synthesis database storage device configured to store the speech synthesis database generated by the speech synthesis database generator.

A media processing device comprising:

• a receiver configured to receive speech data from a relay device;
• a speech data processor configured to store the received speech data in a speech data storage device, wherein the speech data processor is further configured to determine whether a piece of the received speech data corresponds to a piece of previously stored speech data included in the speech data storage device;
• a speech synthesis database generator that generates a speech synthesis database based on speech data stored in the speech data storage device;
• and
• a speech synthesis database storage device configured to store the speech synthesis database generated by the speech synthesis database generator,
• wherein the speech data processor is further configured to determine a first amount of noise contained in the piece of received speech data and a second amount of noise contained in the piece of previously stored speech data that corresponds.

A communication system comprising:

• a relay device connected to a communication network;
• at least two communication terminals connected to the communication network via the relay device, each communication terminal configured to transmit to, and receive speech data from, another communication terminal via the relay device, wherein the at least two communication terminals includes a first communication terminal and a second communication terminal;
• and
• a media processing device connected to the relay device,
• the relay device comprising:
• a transmitter-receiver configured to receive first speech data originating from the first communication terminal and second speech data originating from the second communication terminal, the transmitter-receiver further configured to transmit the received first speech data to the second communication terminal, and transmit the received second speech data to the first communication terminal;
• a data duplicator configured to duplicate speech data;
• and
• a communication controller configured to cause the data duplicator to duplicate at least one of a first piece of speech data included in the received first speech data or a second piece of speech data included in the received second speech data, and to cause the transmitter-receiver to transmit, to the media processing device, the duplicated at least one of the first piece of speech data or the second piece of speech data together with identification information identifying at least one of the first or the second communication terminals as an originating communication terminal, and the media processing device comprising:
• a receiver configured to receive, from the relay device, the duplicated at least one of the first piece of speech data or the second piece of speech data and the identification information;
• a speech data processor configured to receive and store in a speech data storage device the at least one of the first piece of speech data or the second piece of speech data received by the receiver along with the identification information, the speech data processor further configured to measure a first acoustic quality associated with the at least one of the first piece of speech data or the second piece of speech data, and measure a second acoustic quality associated with a piece of previously stored speech data to determine whether the at least one of the first piece of speech data or the second piece of speech data corresponds to the piece of previously stored speech data;
• a speech synthesis database generator configured generate a speech synthesis database for the originating communication terminal based on the speech data stored in the speech data storage device;
• a speech synthesis database storage device configured to store the speech synthesis database generated by the speech synthesis database generator;
• and
• a speech synthesizer configured to execute speech synthesis using the speech synthesis database, the speech synthesis executed by the speech synthesizer in response to receipt of a request for speech synthesis received from the originating communication terminal identified by the identification information as at least one of the first or the second communication terminals.

A relay device for use in a communication system with a first communication terminal and a second communication terminal connected to the communication system via the relay device, the relay device comprising:

• a transmitter-receiver that receives speech data from the first communication terminal and transmits the received speech data to the second communication terminal;
• a data duplicator that duplicates speech data;
• and
• a communication controller that causes the data duplicator to duplicate the speech data received from the first communication terminal via the transmitter-receiver and that causes the transmitter-receiver to transmit the duplicated speech data together with identification information identifying one of the first or the second communication terminals as an originating communication terminal to a media processing device, the media processing device for storage of the duplicated speech data, and generation of a speech synthesis database, the media processing device including a speech data processor configured to determine a first amount of noise contained in the duplicated speech data, and a second amount of noise contained in a piece of previously stored speech data that corresponds to the duplicated speech data.

A relay method for use at a relay device in a communication system with a first communication terminal and a second communication terminal connected to the communication system via the relay device, the method comprising:

• receiving speech data from the first communication terminal and transmitting the received speech data to the second communication terminal;
• duplicating the speech data received in the receiving step;
• transmitting the duplicated speech data together with identification information identifying one of the first or the second communication terminals as an originating communication terminal to a media processing device;
• and
• the media processing device storing the duplicated speech data, generating a speech synthesis database, and determining a first amount of noise contained in the duplicated speech data and a second amount of noise contained in a piece of previously stored speech data that corresponds to the duplicated speech data.

A communication system comprising:

• a relay device connected to a communication network;
• at least two communication terminals connected to the communication network via the relay device, each communication terminal configured to transmit to, and receive speech data from, another communication terminal via the relay device, wherein the at least two communication terminals includes a first communication terminal and a second communication terminal;
• and
• a media processing device connected to the relay device,
• the relay device comprising:
• a transmitter-receiver configured to receive first speech data originating from the first communication terminal and second speech data originating from the second communication terminal, the transmitter-receiver further configured to transmit the received first speech data to the second communication terminal, and transmit the received second speech data to the first communication terminal;
• a data duplicator configured to duplicate speech data;
• and
• a communication controller configured to cause the data duplicator to duplicate at least one of a first piece of speech data included in the received first speech data or a second piece of speech data included in the received second speech data, and to cause the transmitter-receiver to transmit, to the media processing device, the duplicated at least one of the first piece of speech data or the second piece of speech data together with identification information identifying at least one of the first or the second communication terminals as an originating communication terminal, and the media processing device comprising:
• a receiver configured to receive, from the relay device, the duplicated at least one of the first piece of speech data or the second piece of speech data and the identification information;
• a speech data processor configured to receive and store in a speech data storage device the at least one of the first piece of speech data or the second piece of speech data received by the receiver along with the identification information, the speech data processor further configured to determine a first amount of noise contained in the at least one of the first piece of speech data or the second piece of speech data, and a second amount of noise contained in a piece of previously stored speech data that corresponds to the at least one of the first piece of speech data or the second piece of speech data;
• a speech synthesis database generator configured generate a speech synthesis database for the originating communication terminal based on the speech data stored in the speech data storage device;
• a speech synthesis database storage device configured to store the speech synthesis database generated by the speech synthesis database generator;
• and
• a speech synthesizer configured to execute speech synthesis using the speech synthesis database, the speech synthesis executed by the speech synthesizer in response to receipt of a request for speech synthesis received from the originating communication terminal identified by the identification information as at least one of the first or the second communication terminals.