Server-based speech recognition

• US 7,899,670 B1
• Filed: 12/21/2006
• Issued: 03/01/2011
• Est. Priority Date: 12/21/2006
• Status: Active Grant

• Current Assignee: DeliverHealth Solutions LLC
• Original Assignee: eScription, Inc. (Microsoft Corporation)
• Inventors: Zimmerman, Roger S., Young, Jonathan Hood
• Primary Examiner(s): Chawan; Vijay B
• Application Number:
  US11/614,898
• Time in Patent Office: 1,531 Days
• Field of Search: 704/235, 704/270.1, 704/270, 704/275, 704/246, 704/231, 704/257, 704/233, 704/251, 704/250, 704/278, 715/255, 709/250, 709/218, 369/25.01, 379/75
• US Class Current: 704/235
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

A computer program product for transcribing a medical dictation in real-time includes instructions for causing a computer to obtain a user identification from a user at a client via a computer network, the user identification being associated with medical personnel that provides a dictation concerning a patient, load models associated with the user identification, the models being configured to assist in developing a draft transcription of the dictation, receive an audio feed of the dictation from the client, process the audio feed using the models to create a segment of the draft transcription of the dictation, and transmit the segment of the draft transcription to the client.

Claims

1. A computer program product for transcribing a medical dictation in real-time, the computer program product residing on a computer-readable medium and comprising computer-readable instructions for causing a computer to:

obtain a user identification from a client via a computer network, the user identification being associated with and indicative of a speaker that provides the medical dictation concerning a patient;

load models associated with the user identification, the models being configured for use in recognizing words in the medical dictation to develop a draft transcription of the dictation;

receive an audio feed of the dictation from the client;

process the audio feed using the models in real-time to create a segment of the draft transcription of the dictation; and

transmit the segment of the draft transcription to the client in real-time relative to the audio feed.

2. The computer program product of claim 1 further comprising instructions for causing the computer to establish a connection with the client based on an availability of the computer to process the audio feed from the client.

3. The computer program product of claim 1 wherein the instructions for causing the computer to process the audio feed to create the draft transcription of the dictation are configured to cause the computer to:

create the draft transcription of a completed dictation; and

transmit the draft transcription of the completed dictation to the client.

4. The computer program product of claim 1 wherein the instructions for causing the computer to transmit the segment of the draft transcription to the client are configured to cause the computer to transmit the segment of the draft transcription for real-time review at the client.

5. The computer program product of claim 1 further comprising instructions for causing the computer to store the segment of the draft transcription in association with the audio feed.

6. The computer program product of claim 1 further comprising instructions for causing the computer to store all segments of the draft transcription in association with the audio feed.

7. The computer program product of claim 1 further comprising instructions for causing the computer to:

complete the process of creating the segment of the draft transcription;

determine presence of a second segment of the audio feed of the dictation from the client;

process the second segment of the audio feed using the models to create a second segment of the draft transcription of the dictation; and

transmit the second segment of the draft transcription to the client.

8. The computer program product of claim 1 further comprising instructions for causing the computer to receive a sign-out signal from the client upon completion of the dictation.

9. A method of processing a medical record transcription, the method comprising:

obtaining a user identification from a client via a computer network, the user identification being associated with medical personnel that provides a dictation concerning a patient;

loading models associated with the user identification, the models being configured to assist in developing a draft transcription of the dictation;

receiving an audio feed of the dictation from the client;

processing the audio feed using the models to create the draft transcription of the dictation; and

transmitting the draft transcription to the client in real-time relative to the received audio feed.

10. The method of claim 9 further comprising establishing a connection with the client based on a client identification.

11. The method of claim 9 wherein the processing further comprises processing a segment of the audio feed to create a draft transcription of the segment of the audio feed.

12. The method of claim 9 wherein the transmitting further comprises transmitting a segment of the draft transcription to the client.

13. The method of claim 9 further comprising:

refining the draft transcription sent to the client; and

transmitting the refined draft transcription to the client.

14. A method of allocating a medical dictation to a server for transcription, the method comprising:

receiving an identification from a client via a computer network for processing a medical dictation over a network;

selecting an available processor from a plurality of processors in communication over the network;

sending an identification associated with the available processor to the client; and

establishing a link between the client and the available processor such that data is transmitted from the client to the available processor for processing the medical dictation to create a draft transcription of the dictation in real-time.