Idiom handling in voice service systems

• US 20020128845A1
• Filed: 12/13/2001
• Published: 09/12/2002
• Est. Priority Date: 12/13/2000
• Status: Active Grant

• Current Assignee: Regional Resources Ltd (Imperial Metals Corp.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Thomas, Andrew, Hickey, Marianne, Belrose, Guillaume, Hinde, Stephen John
• Granted Patent: US 7,171,361 B2
• Time in Patent Office: Days
• US Class Current: 704/270.1
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 2015/228 of application context

Abstract

The grammar of the speech input to a voice service system is normally specified by the voice service system. However, this can produce problems in respect of idioms, such as dates, which are expressed different ways by different users. To facilitate the handling of idioms, a user is permitted to specify their own idiom grammar which is then used by the voice service system to interpret idioms in speech input from that user. Typically, the normal grammar of speech input is specified by grammar tags used to mark up a voice page script interpreted by a voice page browser; in this case, it will generally be the voice browser that is responsible for employing the user-specified idiom grammar to interpret the corresponding idiom in the speech input by the user. The user-specified grammar can be pre-specified directly to the voice browser by the user or fetched by the browser from a remote location on the fly.

Claims

1. A method of handling idioms in speech input by a user to a voice service system, the grammar of the speech input being normally specified by the voice service system;

said method involving using a user-specified idiom grammar to interpret the corresponding idiom in speech input from that user.

2. A method according to claim 1, wherein the voice service provided by the voice service system is effected by the serving of voice pages in the form of text with embedded voice markup tags, including grammar tags, to a voice browser;

the voice browser interpreting these pages and carrying out speech recognition of user voice input, text to speech conversion to generate voice output, and dialog management;

the voice browser being disposed between a voice page server and the user, and the voice browser retrieving and using said user-specified idiom grammar.

3. A method according to claim 2, wherein the voice browser permanently stores said user-specified idiom grammar and uses it whenever a voice page permits use of a built-in grammar for the idiom concerned.

4. A method according to claim 2, wherein the voice browser fetches said user-specified idiom grammar from a remote location as and when a voice page permits use of a user grammar for the idiom concerned.

5. A method according to claim 2, wherein the voice browser fetches said user-specified idiom grammar from a remote location upon user logon to the browser, the voice browser using the fetched idiom grammar whenever a voice page permits use of a built-in or user grammar for the idiom concerned.

6. A method according to claim 2, wherein the voice browser substitutes said user-specified grammar for a grammar specified in a voice page for that idiom independently of whether the voice page explicitly permits such substitution.

7. A method according to claim 1, wherein said idiom is a date.

8. A voice browser comprising:

a voice-server interface for interfacing with a voice page server to provide user input to the server and to receive voice pages from the server in the form of text with embedded voice markup tags including grammar tags;

an input channel, including a speech recogniser, for interpreting user voice input;

a dialog manager for effecting dialog control on the basis of output from the input channel and voice pages received from the voice page server; and

an output channel, including a text-to-speech converter, for generating voice output from the received voice pages under the control of the dialog manager;

the voice browser further comprising user-grammar functionality for accessing a user-specified idiom grammar and causing the input channel to use that grammar to interpret a corresponding idiom in voice input from that user.

9. A voice browser according to claim 8, wherein the user-grammar functionality includes a data store for permanently storing said user-specified idiom grammar.

10. A voice browser according to claim 9, wherein the user-grammar functionality includes means for determining whether a served voice page containing said idiom permits use of a grammar for that idiom different to any specified in the page itself, the user-grammar functionality being operative to cause the input channel to use the stored user-specified idiom grammar only when permitted by the voice page concerned.

11. A voice browser according to claim 8, wherein the user-grammar functionality includes means for fetching said user-specified grammar from a remote location, the user-grammar functionality being operative to fetch that user-specified idiom grammar for use by the input channel whenever a voice page including the idiom is served to the browser for said user.

12. A voice browser according to claim 8, wherein the user-grammar functionality includes:

means for fetching said user-specified grammar from a remote location, and

means for determining whether a served voice page containing said idiom permits use of a grammar for that idiom different to any specified in the page itself,

the user-grammar functionality being operative to fetch said user-specified grammar for use by the input channel only when a served voice page containing the idiom permits use of a grammar for that idiom different to any specified in the page itself.

13. A voice browser according to claim 8, wherein the user-grammar functionality includes means, responsive to said user connecting to the browser, to fetch said user-specified grammar from a remote location and temporarily store it.

14. A voice browser according to claim 13, wherein the user-grammar functionality includes means for determining whether a served voice page containing said idiom permits use of a grammar for that idiom different to any specified in the page itself, the user-grammar functionality being operative to cause the input channel to use the temporarily-stored user-specified idiom grammar only when permitted by the voice page concerned.

15. A voice browser according to claim 8, wherein the user-grammar functionality voice browser is operative to substitute said user-specified grammar for a grammar specified in a voice page for that idiom independently of whether the voice page explicitly permits such substitution.

16. A voice browser according to claim 8, wherein said idiom is a date.