Method and a system for voice dialling

• US 7,110,948 B1
• Filed: 09/02/1999
• Issued: 09/19/2006
• Est. Priority Date: 09/04/1998
• Status: Expired due to Term

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Mekuria, Fisseha
• Primary Examiner(s): ARMSTRONG, ANGELA A
• Application Number:
  US09/388,609
• Time in Patent Office: 2,574 Days
• Field of Search: 704/242, 704/251, 704/270, 704/275, 379/88.02
• US Class Current: 704/242
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: H04M 1/271 controlled by voice recognition

Abstract

In a speech recognition system of a mobile telephone the words are organized in a trellis structure. Thus, at each instant the speech recognition system only needs to search a limited part of the entire vocabulary. Such an arrangement solves the problem of having to search many words at a time, which is time consuming and imposes a high computational load on the system, and will therefore significantly increase the accuracy for the speech recognition system.

Claims

1. A speech recognition system in a mobile telephone, the speech recognition system comprising:

a stored vocabulary, wherein words in the vocabulary are arranged in a trellis structure comprising a plurality of different groups of words, and

a word group selection system for enabling a user to speak via voice commands to select at least a first of said plurality of different groups of words, said first group of words being selected based upon at least a word spoken by the user, so that a limited number of groups of the entire vocabulary, less than said plurality, is searched for a word during subsequent speech recognition processes in the mobile telephone after selection of at least the first of said plurality of groups of words.

2. A system according to claim 1, characterized in that the vocabulary is arranged in a tree structure.

3. A system according to claim 1, characterized by

means for outputting the words that the system is set to recognize at a particular moment.

4. A system according to claim 3, characterized in that said means is a voice prompter.

5. A system according to claim 1, characterized by

means for automatically generating a new group if the number of words in one group exceeds a certain, pre-set threshold value.

11. The speech recognition system of claim 1, further comprising an automatic word group generation system for automatically generating new groups of words for storage in said vocabulary when a number of words in or at a particular location in the vocabulary exceeds a predetermined threshold value.

6. A method of speech recognition comprising:

providing a speech recognition system of a mobile telephone comprising a stored vocabulary, wherein the words in the stored vocabulary are arranged in a trellis structure comprising a plurality of different groups of words,

providing a word group selection system for enabling a user to speak via voice commands to select at least a first of said plurality of different groups of words, said first group of words being selected based upon at least a word spoken by the user, so that only one group or a limited number of groups of the entire vocabulary less than said plurality is searched for a word during certain subsequent speech recognition processes in the mobile telephone after selection of at least the first of said plurality of groups of words.

7. A method according to claim 6, characterized in that the vocabulary is arranged in a tree structure.

8. A method according to claim 6, characterized in that the available words that the system is set to recognize at a particular moment is output from the system.

9. A method according to claim 8, characterized in that the available words are generated by a voice prompter.

10. A method according to claim 6, characterized in that a new group automatically is generated if the number of words in one group exceeds a certain, pre-set threshold value.

12. A speech recognition system in a mobile telephone, the speech recognition system comprising:

means for storing a word vocabulary in trellis tree structure, wherein words in the vocabulary are arranged in a plurality of different groups of words,

word group selection means for enabling a user to speak via voice commands into the mobile telephone to select a first of said plurality of different groups of words, said first group of words being selected based upon at least a word spoken by the user, and

speech recognition means for comparing input speech from a user to words in said selected first group of words, so that comparing of the input speech is performed relative to said selected first group of words prior to comparing the input speech with other of the plurality of different groups of words so that a limited number of groups of the entire vocabulary is searched via said comparing during speech recognition processes.

13. The speech recognition system of claim 12, further comprising automatic word group generation means for automatically generating new groups of words for storage in said vocabulary when a number of words in or at a particular location in the vocabulary exceeds a predetermined threshold value.