Method and system for acquiring information with voice input

• US 7,190,770 B2
• Filed: 07/29/2002
• Issued: 03/13/2007
• Est. Priority Date: 02/18/2002
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Matsuda, Junichi, Ando, Haru, Nejime, Yoshito
• Primary Examiner(s): Escalante; Ovidio
• Application Number:
  US10/206,231
• Publication Number:
  US 20030156689A1
• Time in Patent Office: 1,688 Days
• Field of Search: 379/88.01, 379/88.02, 379/88.03, 379/88.04, 704/231, 704/246, 704/251, 704/270
• US Class Current: 379/88.01
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML

Abstract

A user-requested command with both visual and hearing information is indicated. Additionally, with a user friendly method, the user understands what content to input. The user-requested command is expressed in the form of a template sentence. Template part in the template sentence is vocalized, and a slot area in the template sentence is expressed using a sound or voice. The user inputs his or her voice to be input to the slot area.

Claims

1. A method for acquiring information with voice inputs, comprising the steps of:

outputting a template sentence including a slot area as sound data, wherein the template sentence except the slot area is voice-synthesized and the slot area comprises a sound of a predetermined time length that is not voice-synthesized;

receiving voice data including a word corresponding to the slot area of the output template sentence;

converting the received voice data to a character string through a voice recognition process;

acquiring a character string corresponding to the word from the character string obtained in the step of converting; and

wherein said sound is superposed by a sample of a word to be input to said slot area by using a voice.

2. The method for acquiring information with voice input according to claim 1, wherein said slot area has a category as an attribute, and when the category of the slot area matches a category of the word in said character string, the word corresponding to the slot area is acquired.

3. The method for acquiring information with inputs according to claim 1, wherein the sound of the predetermined time length comprises chime or beep.

4. A method for acquiring information with voice input, comprising the steps of:

receiving an instruction to select one of a plurality of template sentences including slot areas, the template sentences being stored in a memory device;

producing sound data including voice synthesization of the selected template sentence wherein the slot area of the selected template sentence comprises a sound of a predetermined time length that is not a synthesized word;

transmitting the produced sound data;

receiving voice data including a word corresponding to the slot area of the template sentence;

converting the received voice data to a character string through a voice recognition process;

acquiring a character string corresponding to the word from the character string obtained in the step of converting; and

wherein said sound is superposed by a sample of a word to be input to said slot area by using a voice.

5. The method for acquiring information with voice input according to claim 4, wherein said slot area has a category as an attribute, and when the category of the slot area matches a category of the word in said character string, the word corresponding to the slot area is acquired.

6. The method for acquiring information with inputs according to claim 4, wherein the sound of the predetermined time length comprises chime or beep.

7. The method for acquiring information with voice inputs, comprising the steps of:

outputting a template sentence including a slot area as sound data, wherein the template sentence except the slot area is voice-synthesized and the slot area comprises a sound of a predetermined time length that is not voice-synthesized;

receiving voice data including a word corresponding to the slot area of the output template sentence;

converting the received voice data to a character string through a voice recognition process;

acquiring a character string corresponding to the word from the character string obtained in the step of converting; and

wherein a time length of the sound replacing said slot area varies according to said slot area.

8. The method for acquiring information with voice input, comprising the steps of:

receiving an instruction to select one of a plurality of template sentences including slot areas, the template sentences being stored in a memory device;

producing sound data including voice synthesization of the selected template sentence wherein the slot area of the selected template sentence comprises a sound of a predetermined time length that is not a synthesized word;

transmitting the produced sound data;

receiving voice data including a word corresponding to the slot area of the template sentence;

converting the received voice data to a character string through a voice recognition process;

acquiring a character string corresponding to the word from the character string obtained in the step of converting; and

wherein a time length of the sound replacing said slot area varies according to said slot area.

9. A system for acquiring information with voice input, comprising:

storing means for storing a plurality of template sentences including slot areas, and sound data corresponding to the slot areas;

voice synthesizing means for converting the template sentence including the slot area to synthesized sound data, wherein the slot area comprises the sound data which is not a voice-synthesized word and which is extracted according to time length data that corresponds to the slot area of the template sentence;

transmitting means for transmitting the synthesized sound data;

receiving means for receiving voice data including a word corresponding to the slot area of the template sentence;

voice recognition means for converting the voice data received by the receiving means to a character string;

word acquiring means for acquiring a character string corresponding to the slot area in the transmitted template sentence from the character string obtained by the voice recognition means; and

wherein said sound data corresponding to the slot area is superposed by a sample word to be input to the slot areas by using a voice.

10. The system for acquiring information according to claim 9, wherein a word dictionary that registers words corresponding to said slot areas is provided, and said word acquiring means acquires words corresponding to the slot areas by collating a word in said character string and the words registered in the word dictionary.

11. The system for acquiring information according to claim 10, wherein said sound data corresponding to the slot area has a word category, and said word dictionary has data on word categories, and wherein said word acquiring means acquires a word corresponding the slot area by collating the word category of said slot area and the word category of the word in said character string.