# US 2004/0083103 A1 · Speech recognition method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2004/0083103 A1
- Document ID: US20040083103A1
- Status: Published application
- Kind: A1
- Application number: 10/687,214
- Filed: 2003-10-15
- Published: 2004-04-29
- Earliest claimed: 1996-12-06
- Assignees: Sensory, Incorporated
- Inventors: Todd F. Mozer; Forrest S. Mozer
- Web page: https://patentgrep.com/patent/US20040083103A1

## CPC

- G04G 21/06: using voice
- G10L 15/285: {Memory allocation or algorithm optimisation to reduce hardware requirements}
- G04G: ELECTRONIC TIME-PIECES
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Embodiments of the present invention include a speech recognition method. In one embodiment, the method includes receiving from an external system first recognition information to recognize a first plurality of words in a first system, programming the first system with the first recognition information to recognize the first plurality of words, generating first recognition results in response to receiving at least one of the first plurality of words in the first system, receiving from the external system second recognition information to recognize a second plurality of words, wherein the second recognition information is selected based on the first recognition results, and programming the first system with the second recognition information to recognize a second plurality of words.

## Claims

### Claim 1

A method comprising the steps of:
retrieving a first set of recognition programming information associated with a first recognition word set from a remote source external to a first system;
programming the first system with said retrieved recognition programming information to recognize a first set of words in the first recognition word set, and in accordance therewith, generating a first set of recognition results responsive to said first set of words;
in response to the first set of recognition results, selectively retrieving from said remote external source a second set of recognition programming information associated with a second recognition word set, wherein the selectively retrieved second set of recognition programming is selected based on the first set of recognition results; and
reprogramming the first system with said retrieved second set of recognition programming information to recognize a second set of words in the second recognition word set.

### Claim 2 (depends on 1)

The method of claim 1 wherein the first system is connected to the remote source over an internet.

### Claim 3 (depends on 1)

The method of claim 1 wherein the first system is connected to the remote source over an intranet.

### Claim 4 (depends on 1)

The method of claim 1 wherein the remote source is a server.

### Claim 5 (depends on 1)

The method of claim 1 wherein the first system receives word sets from a web site.

### Claim 6 (depends on 1)

The method of claim 1 wherein recognition is performed using speaker-independent speech recognition.

### Claim 7 (depends on 1)

The method of claim 1 wherein the recognition programming information comprises neural network weights.

### Claim 8

A method comprising:
receiving from an external system first recognition information to recognize a first plurality of words in a first system;
programming the first system with the first recognition information to recognize the first plurality of words;
generating first recognition results in response to receiving at least one of the first plurality of words in the first system;
receiving from the external system second recognition information to recognize a second plurality of words, wherein the second recognition information is selected based on the first recognition results; and
programming the first system with the second recognition information to recognize a second plurality of words.

### Claim 9 (depends on 8)

The method of claim 8 further comprising using a speaker to prompt a user to input at least one of the first plurality of words.

### Claim 10 (depends on 8)

The method of claim 8 further comprising using a speaker to prompt a user to input at least one of the second plurality of words.

### Claim 11 (depends on 8)

The method of claim 8 further comprising receiving data from the external system.

### Claim 12 (depends on 11)

The method of claim 11 wherein the data is audio data.

### Claim 13 (depends on 11)

The method of claim 11 wherein the data is video data.

### Claim 14 (depends on 8)

The method of claim 8 wherein the first system is connected to the external system over an internet.

### Claim 15 (depends on 8)

The method of claim 8 wherein the first system is connected to the external system over an intranet.

### Claim 16 (depends on 8)

The method of claim 8 wherein the external system is a server.

### Claim 17 (depends on 8)

The method of claim 8 wherein the first system receives the first and second plurality of words from a web site.

### Claim 18 (depends on 9)

The method of claim 9 further comprising using a speaker to prompt a user to input at least one of the second plurality of words.

### Claim 19 (depends on 18)

The method of claim 18 further comprising receiving data from the external system.

### Claim 20 (depends on 19)

The method of claim 19 wherein the data is audio data.

### Claim 21 (depends on 19)

The method of claim 19 wherein the first system is connected to the external system over an internet.

### Claim 22 (depends on 19)

The method of claim 19 wherein the first system is connected to the external system over an intranet.

### Claim 23 (depends on 21)

The method of claim 21 wherein the external system is a server.

### Claim 24 (depends on 21)

The method of claim 21 wherein the first system receives the first and second plurality of words from a web site.

### Claim 25 (depends on 22)

The method of claim 22 wherein the external system is a server.

### Claim 26 (depends on 22)

The method of claim 22 wherein the first system receives the first and second plurality of words from a web site.

### Claim 27 (depends on 8)

The method of claim 8 wherein recognition is performed using a speaker-independent speech recognition.

### Claim 28 (depends on 8)

The method of claim 8 wherein recognition programming information comprises neural network weights.

### Claim 29 (depends on 8)

The method of claim 8 wherein the first system includes means for recognizing speech and means for coupling the first system to the external system.

### Claim 30 (depends on 8)

The method of claim 8 wherein the first system is a base unit.

### Claim 31 (depends on 8)

The method of claim 8 wherein the first system includes a recognition engine.

### Claim 32 (depends on 31)

The method of claim 31 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 33 (depends on 8)

The method of claim 8 wherein the first system is a local computer and the external system is an internet web site, and the first and second recognition information are downloaded from the internet web site to the local computer to recognize the first and second plurality of words.

### Claim 34

A method comprising:
supplying a first system with first information from an external source, the first information for recognizing a spoken utterance from a first limited set of candidate utterances;
programming the first system with the first information to recognize the first limited set of candidate utterances;
generating first results in response to receiving a spoken utterance from the first limited set of candidate utterances in the first system;
supplying the first system with second information from the external source, the second information for recognizing a spoken utterance from a second limited set of candidate utterances, wherein the second information is selected based on the first results; and
programming the first system with the second information to recognize the second limited set of candidate utterances.

### Claim 35 (depends on 34)

The method of claim 34 wherein the first system includes a recognition engine.

### Claim 36 (depends on 35)

The method of claim 35 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 37 (depends on 34)

The method of claim 34 wherein the first system is a base unit.

### Claim 38 (depends on 37)

The method of claim 37 wherein the base unit is an internet access device.

### Claim 39 (depends on 34)

The method of claim 34 wherein the first system is connected to the external source over an internet.

### Claim 40 (depends on 34)

The method of claim 34 wherein the first system is connected to the external source over an intranet.

### Claim 41 (depends on 34)

The method of claim 34 wherein the external source is a server.

### Claim 42 (depends on 41)

The method of claim 41 wherein the first system receives the first and second information from a web site.

### Claim 43 (depends on 34)

The method of claim 34 wherein the first system is a local computer and the external source is an internet web site, and the first and second information are downloaded from the internet web site to the local computer for recognizing the first and second limited set of candidate utterances.

### Claim 44

A method comprising:
downloading first information to a first computer from a server, the first information including data to recognizing an utterance from a first limited set of candidate utterances;
programming the first computer with the first information to recognize the first limited set of candidate utterances;
receiving a spoken utterance from the first limited set of candidate utterances in the first computer;
generating first recognition results in response to receiving the spoken utterance;
downloading second information to the first computer from the server, the second information including data to recognizing an utterance from a second limited set of candidate utterances, wherein the second information is selected based on the first recognition results; and
programming the first computer with the second information to recognize the second limited set of candidate utterances.

### Claim 45 (depends on 44)

The method of claim 44 wherein the first computer includes a recognition engine.

### Claim 46 (depends on 45)

The method of claim 45 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 47 (depends on 44)

The method of claim 44 wherein the first computer is connected to the server over an internet.

### Claim 48 (depends on 44)

The method of claim 44 wherein the first computer is connected to the server over an intranet.

### Claim 49 (depends on 44)

The method of claim 44 wherein the first computer is connected to the server over a local network.

### Claim 50 (depends on 44)

The method of claim 44 wherein the first computer receives the first and second information from a web site.

### Claim 51 (depends on 44)

The method of claim 44 wherein the server is an internet web site, and the first and second information are downloaded from the internet web site to the first computer for recognizing the first and second limited set of candidate utterances.

### Claim 52 (depends on 44)

The method of claim 44 further comprising receiving data from the server.

### Claim 53 (depends on 52)

The method of claim 52 wherein the data is audio data.

### Claim 54 (depends on 52)

The method of claim 52 wherein the data is video data.

### Claim 55 (depends on 44)

The method of claim 44 further comprising prompting a user to input a spoken utterance from the first limited set of candidate utterances.

### Claim 56 (depends on 55)

The method of claim 55 further comprising prompting a user to input a spoken utterance from the second limited set of candidate utterances.

### Claim 57 (depends on 44)

The method of claim 44 wherein recognition is performed using speaker-independent speech recognition.

### Claim 58 (depends on 57)

The method of claim 57 wherein the first and second information comprise neural network weights.

### Claim 59

A computer-readable storage medium including software for performing a method, the method comprising:
receiving in a first system from a second system first recognition information to recognize a first plurality of words;
programming the first system with the first recognition information to recognize the first plurality of words;
generating first recognition results in response to receiving at least one of the first plurality of words in the first system;
receiving from the second system second recognition information to recognize a second plurality of words, wherein the second recognition information is selected based on the first recognition results; and
programming the first system with the second recognition information to recognize a second plurality of words.

### Claim 60 (depends on 59)

The method of claim 59 further comprising prompting a user to input at least one of the first plurality of words.

### Claim 61 (depends on 60)

The method of claim 60 further comprising prompting a user to input at least one of the second plurality of words.

### Claim 62 (depends on 59)

The method of claim 59 further comprising receiving data from the second system.

### Claim 63 (depends on 59)

The method of claim 59 wherein the first system is connected to the second system over an internet.

### Claim 64 (depends on 59)

The method of claim 59 wherein the first system is connected to the second system over an intranet.

### Claim 65 (depends on 59)

The method of claim 59 wherein the first system is a local computer.

### Claim 66 (depends on 65)

The method of claim 65 wherein the second system is a server.

### Claim 67 (depends on 59)

The method of claim 59 wherein the first system receives the first and second plurality of words from a web site.

### Claim 68 (depends on 59)

The method of claim 59 wherein recognition is performed using speaker-independent speech recognition.

### Claim 69 (depends on 59)

The method of claim 59 wherein recognition programming information comprises neural network weights.

### Claim 70 (depends on 59)

The method of claim 59 wherein the first system includes a recognition engine.

### Claim 71 (depends on 70)

The method of claim 70 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 72 (depends on 59)

The method of claim 59 wherein the first system is a local computer and the external system is an internet web site, and the first and second recognition information are downloaded from the internet web site to the local computer to recognize the first and second plurality of words.

### Claim 73

A computer-readable storage medium including software for performing a method, the method comprising:
supplying a first system with first information from an external source, the first information for recognizing a spoken utterance from a first limited set of candidate utterances;
programming the first system with the first information to recognize the first limited set of candidate utterances;
generating first results in response to receiving a spoken utterance from the first limited set of candidate utterances in the first system;
supplying the first system with second information from the external source, the second information for recognizing a spoken utterance from a second limited set of candidate utterances, wherein the second information is selected based on the first results; and
programming the first system with the second information to recognize the second limited set of candidate utterances.

### Claim 74 (depends on 73)

The method of claim 73 wherein the first system includes a recognition engine.

### Claim 75 (depends on 74)

The method of claim 74 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 76 (depends on 73)

The method of claim 73 wherein the first system is a base unit.

### Claim 77 (depends on 76)

The method of claim 76 wherein the base unit is an internet access device.

### Claim 78 (depends on 73)

The method of claim 73 wherein the first system is connected to the external source over an internet.

### Claim 79 (depends on 73)

The method of claim 73 wherein the first system is connected to the external source over an intranet.

### Claim 80 (depends on 73)

The method of claim 73 wherein the external source is a server.

### Claim 81 (depends on 80)

The method of claim 80 wherein the first system receives the first and second information from a web site.

### Claim 82 (depends on 73)

The method of claim 73 wherein the first system includes means for recognizing speech and means for coupling the first system to the external source.

### Claim 83 (depends on 73)

The method of claim 73 wherein the first system is a local computer and the external source is an internet web site, and the first and second information are downloaded from the internet web site to the local computer for recognizing the first and second limited set of candidate utterances.

### Claim 84

A computer-readable storage medium including software for performing a method, the method comprising:
downloading first information to a first computer from a server, the first information including data to recognizing an utterance from a first limited set of candidate utterances;
programming the first computer with the first information to recognize the first limited set of candidate utterances;
receiving a spoken utterance from the first limited set of candidate utterances in the first computer;
generating first recognition results in response to receiving the spoken utterance;
downloading second information to the first computer from the server, the second information including data to recognizing an utterance from a second limited set of candidate utterances, wherein the second information is selected based on the first recognition results; and
programming the first computer with the second information to recognize the second limited set of candidate utterances.

### Claim 85 (depends on 84)

The method of claim 84 wherein the first computer includes a recognition engine.

### Claim 86 (depends on 85)

The method of claim 85 wherein the recognition engine is a software program running in a general purpose microprocessor.

### Claim 87 (depends on 84)

The method of claim 84 wherein the first computer is connected to the server over an internet.

### Claim 88 (depends on 84)

The method of claim 84 wherein the first computer is connected to the server over an intranet.

### Claim 89 (depends on 84)

The method of claim 84 wherein the first computer is connected to the server over a local network.

### Claim 90 (depends on 84)

The method of claim 84 wherein the first computer receives the first and second information from a web site.

### Claim 91 (depends on 84)

The method of claim 84 wherein the server is an internet web site, and the first and second information are downloaded from the internet web site to the first computer for recognizing the first and second limited set of candidate utterances.

### Claim 92 (depends on 84)

The method of claim 84 further comprising receiving data from the server.

### Claim 93 (depends on 92)

The method of claim 92 wherein the data is audio data.

### Claim 94 (depends on 92)

The method of claim 92 wherein the data is video data.

### Claim 95 (depends on 94)

The method of claim 94 further comprising prompting a user to input a spoken utterance from the first limited set of candidate utterances.

### Claim 96 (depends on 95)

The method of claim 95 further comprising prompting a user to input a spoken utterance from the second limited set of candidate utterances.

### Claim 97 (depends on 84)

The method of claim 84 wherein recognition is performed using speaker-independent speech recognition.

### Claim 98 (depends on 97)

The method of claim 97 wherein the first and second information comprise neural network weights.

### Claim 99 (depends on 1)

The method of claim 1 wherein the wherein the first system is connected to the remote source over a wireless connection.

### Claim 100 (depends on 8)

The method of claim 8 wherein the first system is connected to the external system over a wireless connection.

### Claim 101 (depends on 34)

The method of claim 34 wherein the first system is connected to the external source over a wireless connection.

### Claim 102 (depends on 59)

The method of claim 59 wherein the first system is connected to the second system over a wireless connection.

### Claim 103 (depends on 73)

The method of claim 73 wherein the first system is connected to the external source over a wireless connection.

### Claim 104 (depends on 84)

The method of claim 84 wherein the first computer is connected to the server over a wireless connection.
