Instantaneous context switching for speech recognition systems
First Claim
Patent Images
1. In a speech recognizer for multiple contexts and a plurality of users, an instantaneous context switching system, comprising:
- a memory coupled to a data processing means, a speech input means coupled to the plurality of users for speech recognition purposes and an output character string utilization device;
a first context partition in said memory including a first plurality of words, each said first plurality of words including a second plurality of phonemes;
a second context partition in said memory including a third plurality of words, each said third plurality of words including a fourth plurality of phonemes;
a pattern matching partition in said memory including a fifth plurality of phoneme pattern matching data units for sharing by both contexts;
a first pointer map including a second plurality of pointers, each respective one of said second plurality of phonemes related to a respective one of said of fifth pattern matching data units;
a second pointer map including a fourth plurality of pointers, each respective one of said fourth plurality of phonemes related to a respective one of said fifth plurality of pattern matching data units; and
selection means coupled to said memory either for selecting said first context partition and said first pointer map or alternately for selecting said second context partition and said second pointer map, for converting speech input information received from said speech input means into character string information output to said character string utilization device using the shared phoneme pattern matching data units for both contexts;
whereby, a context of a speech recognition application can be instantaneously changed without loading new pattern matching data units in said memory.
0 Assignments
0 Petitions
Accused Products
Abstract
An instantaneous context switching speech recognition system is disclosed which enables a speech recognition application to be changed without loading new pattern matching data into the system. Selectable pointer maps are included in the memory of the system which selectively change the relationship between words and phonemes between a first application context and the pattern matching logic to a second application context and the pattern matching logic.
158 Citations
19 Claims
-
1. In a speech recognizer for multiple contexts and a plurality of users, an instantaneous context switching system, comprising:
-
a memory coupled to a data processing means, a speech input means coupled to the plurality of users for speech recognition purposes and an output character string utilization device; a first context partition in said memory including a first plurality of words, each said first plurality of words including a second plurality of phonemes; a second context partition in said memory including a third plurality of words, each said third plurality of words including a fourth plurality of phonemes; a pattern matching partition in said memory including a fifth plurality of phoneme pattern matching data units for sharing by both contexts; a first pointer map including a second plurality of pointers, each respective one of said second plurality of phonemes related to a respective one of said of fifth pattern matching data units; a second pointer map including a fourth plurality of pointers, each respective one of said fourth plurality of phonemes related to a respective one of said fifth plurality of pattern matching data units; and selection means coupled to said memory either for selecting said first context partition and said first pointer map or alternately for selecting said second context partition and said second pointer map, for converting speech input information received from said speech input means into character string information output to said character string utilization device using the shared phoneme pattern matching data units for both contexts; whereby, a context of a speech recognition application can be instantaneously changed without loading new pattern matching data units in said memory. - View Dependent Claims (2)
-
-
3. In a speech recognizer for multiple contexts and a plurality of users, a method for instantaneous context switching comprising the steps of:
-
loading a first plurality of words into a first context partition in a memory in said speech recognizer, said first plurality of words including a second plurality of phonemes; loading a third plurality of words in a second context partition in said memory, each said third plurality of words including a fourth plurality of phonemes; loading a fifth plurality of phonemes pattern matching data units in a pattern matching partition in said memory for sharing by both contexts;
loading a first pointer map in said memory including, a second plurality of pointers, each respective one of said second plurality of phonemes related to a respective one of said fifth plurality of shared pattern matching data units;loading a second pointer map into said memory, including a fourth plurality of pointers, each respective one of said fourth plurality of phonemes related to a respective one of said fifth plurality of said shared pattern matching data units; providing a voice input for speech recognition purpose from one of a plurality of users of the speech recognizer; and selecting either said first context partition and said first pointer map or alternately selecting said second context partition and said second pointer map using the shared phoneme pattern matching data units for both contexts; whereby, a context of a speech recognition application can be instantaneously changed without loading new pattern matching data units in said memory. - View Dependent Claims (4)
-
-
5. A speech recognition system for multiple contexts and multiple users comprising:
-
a memory containing a plurality of contexts, each context containing a plurality of words, each word having a plurality of phonemes; a plurality of phoneme pattern matching data units stored in the memory for sharing by all contexts; a plurality of pointer maps stored in the memory, each map linking all phonemes in a context to the phoneme pattern data units; a speech input adapter for receiving speech input from a user; selection means coupled to the memory for selecting a context and a pointer map in response to a speech input of one or more words in a context, the pointer map coupling the phonemes for each word in the context to the phoneme pattern matching data units to recognize and convert the speech input into character string information using the shared phoneme pattern matching data units for all contexts; and an output adapter for presenting the recognized speech. - View Dependent Claims (6)
-
-
7. A speech recognition system comprising:
-
speech input means for generating a series of vector quantization values (VQ) indicative of speech input; means for generating multiple word grammars as contexts, each context related to a user application; a speech recognition unit for matching word sequences in a context to the series of VQ values in the speech input comprising; means coupling the speech input means, a processor, and an output device to a memory; a plurality of the contexts stored in the memory, each context containing a plurality of words represented by phonemes; a phoneme pattern matching unit stored in the memory for sharing by all contexts in the speech recognition unit; a plurality of pointer maps stored in the memory, each pointer map coupled to a different context, each pointer in a map relating a phoneme in a word to a phoneme in the pattern matching unit; stored program instructions in the memory for operating the processor to match the VQ values in the speech input to the words stored in the context and generating character strings representative of the speech input as an output in the output device; and means for instantaneously switching the speech recognition unit from one context to another context without changing the phoneme pattern matching units. - View Dependent Claims (8, 9, 10, 11, 12)
-
-
13. In a continuous speech recognition system for recognizing input speech in different contexts, a front end for receiving the speech related to the contexts, a speech recognition unit including a processor and a memory for generating character strings representative of the speech input, a method of speech recognition comprising the steps of:
-
storing in the memory a plurality of contexts of words, each word represented by at least one phoneme; storing in the memory a plurality of phoneme pattern matching units for sharing by all contexts; storing in the memory pointer maps, each pointer map relating phonemes in the words to phonemes in the phoneme pattern matching units; storing in the memory program instructions for operating the processor; operating the processor using the stored program instruction to match the speech input to the words in the context using the shared phoneme pattern matching units; generating character strings as an output from the system from the matched speech input and words in the context; and
instantaneously switching the system to another context stored in the memory without changing the phoneme pattern matching units. - View Dependent Claims (14, 15, 16, 17, 18, 19)
-
Specification