Two-way speech recognition and dialect system

• US 6,853,971 B2
• Filed: 06/21/2002
• Issued: 02/08/2005
• Est. Priority Date: 07/31/2000
• Status: Active Grant

• Current Assignee: Ovonyx Memory Technology, LLC
• Original Assignee: Micron Technology, Inc.
• Inventors: Taylor, George W.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Nolan, Daniel A
• Application Number:
  US10/178,182
• Publication Number:
  US 20020161580A1
• Time in Patent Office: 963 Days
• Field of Search: 704/235, 704/200.1, 704/230, 704/201, 704/270.1
• US Class Current: 704/10
• CPC Class Codes: G10L 15/005 Language recognition
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A speech-to-text conversion system. The two-way speech recognition and dialect system comprises a computer system, an attached microphone assembly, and speech-to-text conversion software. The two-way speech recognition and dialect system includes a database of dialectal characteristics and queries a user to determine their likely dialect. The system uses this determination to reduce the time for the system to reliably transcribe a user'"'"'s speech into text and to anticipate dialectal word usage. In another embodiment of the invention, the two-way speech recognition and dialect system is capable of transcribing the speech of multiple speakers while distinguishing between the different speakers and identifying the text belonging to each speaker.

Claims

1. A system for transcribing spoken words from multiple users to text, the system comprising:

at least one audio transducer that receives spoken words and generates speech signals corresponding thereto;

at least one user interface through which users can receive and send signals;

at least one data structure containing word data correlating text representations of words to speech signals wherein the at least one data structure includes dialect parameter data that can be used to recognize selected dialects corresponding to selected users; and

a processor that receives the speech signals from the at least one audio transducer wherein the processor initially sends signals to the users via the at least one user interface asking the users questions indicative of the users'"'"' dialects to thereby induce the users to provide answer signals to the processor that are indicative of dialect parameters and wherein the processor thereafter compares the received speech signals to the word data in the at least one data structure to produce text representations corresponding to the spoken words and associates received speech signals from the users with the corresponding text representation and stores the associations and wherein the processor analyses the received speech signals and determines distinguishing dialectal characteristics of the received speech signals and wherein the processor uses the distinguishing dialectal characteristics to identify the text representations associated with the received speech signals corresponding to each user.

2. The system of claim 1 wherein the processor displays the text representation corresponding to the spoken words of each user in a distinguishable manner via the at least one user interface.

3. The system of claim 2, wherein the distinguishable manner comprises displaying the representative text via the at least one user interface in different colors.

4. The system of claim 1, wherein the data structure further includes a plurality of text documents corresponding to each dialect and wherein the system presents the text document corresponding to the dialect determined for each of the multiple users so that the users can read the text documents aloud and wherein the system compares the text generated from the readings of the text documents to the actual content of the text documents and performs a self-correction process to improve the accuracy of the speech-to-text transcription and wherein each text document is adapted to minimize the time required for the system to perform the self-correction process.