Distributed dictation/transcription system

• US 8,150,689 B2
• Filed: 12/19/2008
• Issued: 04/03/2012
• Est. Priority Date: 12/21/2007
• Status: Active Grant

• Current Assignee: nVoq, Inc.
• Original Assignee: nVoq, Inc.
• Inventors: Beach, Richard, Butler, Christopher, Ford, Jon, Marquette, Brian, Omland, Christopher
• Primary Examiner(s): McFadden, Susan
• Application Number:
  US12/339,366
• Publication Number:
  US 20090177470A1
• Time in Patent Office: 1,201 Days
• Field of Search: 704/235
• US Class Current: 704/235
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: H04L 67/1004 Server selection for load balancing
• CPC Class Codes: H04L 67/1027 Persistence of sessions during load balancing
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54
• CPC Class Codes: H04L 67/568 Storing data temporarily at an intermediate stage, e.g. caching

Abstract

A distributed dictation/transcription system is provided. The system provides a client station, dictation manager, and dictation server networked such that the dictation manager can select a dictation server to transcribe audio from the client station. The dictation manager selects one of a plurality of dictation servers based on conventional load balancing as well as on a determination of which of the dictation servers may already have a user profile uploaded. Moreover, while selecting a dictation server and/or uploading a profile, the user or client at the client station may begin dictating, which audio would be stored in a buffer of dictation manager until a dictation server was selected and/or available. The user would receive in real time or near real time a display of the textual data that may be corrected by the user. The corrective textual data may be transmitted back to the dictation manager to update the user profile.

Claims

1. A dictation manager having a network connection to at least one client station and at least one dictation server, comprising:

at least one network data port to receive audio signals from a client station, to transmit the received audio signals to a dictation server, to receive textual data from the dictation server where the textual data comprises data from the audio signal transformed into textual data by the dictation server, and to transmit textual data to the at least one client station;

a dictation server selector to select one of the at, least one dictation servers from which the dictation manager transrhits the received audio signal and receives the textual data;

at least one memory containing a user profile, an indication of whether the user profile has been uploaded to and maintained by any of the at least one dictation server, an indication of which of the at least one dictation server to which the user profile has been uploaded, and a buffer to cache the received audio signal;

wherein if the dictation server selector determines that the user profile has, been uploaded previously and prior to transmitting the received audio to one of the at least one dictation servers, the dictation server selector selects as the one of the at least one dictation server a dictation server to which the user profile has already been uploaded;

wherein if the dictation server selector determines that the user profile has not been uploaded, the dictation server selector selects any one of the at least one dictation server and uploads the user profile; and

wherein the dictation manager stores the received audio signal in the buffer and transmits the audio signal from the buffer once the dictation server is selected by the dictation server selector, such that the textual data is transmitted to the at least one client station in at least one of real-time or near real-time.

2. The dictation manager of claim 1 wherein the at least one network data port comprises a plurality of data ports configured to receive and transmit audio signals and data signals.

3. The dictation manager of claim 2 wherein the audio signal is stored in the buffer while the dictation server selector is selecting one of the at least one dictation server.

4. The dictation manager of claim 1 wherein the dictation server selector further comprises a load balancer.

5. The dictation manager of claim 4 wherein the dictation server selector Selects a dictation server based on a determination of the load balancer only if the user profile has not been previously uploaded.

6. The dictation manager of claim 5 wherein the determination is based at least one of queue length, average time to process, process load, and memory.

7. The dictation manager of claim 1 further comprising a user profile corrector wherein the dictation manager receives corrective textual data from the client station and the user profile corrector updates the user profile based on the corrective textual data.

8. A method of distributed dictation and transcription performed using at least one processor associated with a dictation manager, the method comprising the steps of:

receiving an audio signal from a user operating a client station;

identifying a user profile stored in a memory of the dictation manager associated with the user of the received audio signal;

determining whether the identified user profile Of the user has been previously provided to and is still retained by one of a plurality of dictation servers coupled to the dictation manager based on a previously received audio signal;

selecting from the plurality of dictation servers the one to which the user profile has already been provided and is still retained by;

causing the audio signal to be converted into a textual data signal by transmitting the received audio signal to the selected dictation server and receiving from the selected dictation server transcription textual data; and

transmitting transcription textual data to the client station in at least one of real-time or near real-time.

9. The method of claim 8 wherein the step of selecting the dictation server also comprises balancing a load between dictation servers,

10. The method of claim 8 wherein the step of determining whether the identified user profile has been previously provided comprises determining at least two dictation servers of the plurality of dictation servers have been previously provided with the identified user profile and the step of selecting the dictation server further comprises selecting one of the at least two dictation servers to balance a load between the at least two dictation servers.

11. The method of claim 8 wherein the step of determining whether the identified user profile has been previously provided to any of the plurality of dictation servers determines the identified user profile has not been previously provided and the step of selecting the one dictation server from the plurality of dictation servers comprises balancing a load between the plurality of dictation servers and further comprising the step of transmitting the identified user profile to the selected one of the plurality of dictation servers.

12. The method of claim 8 further comprising the step of buffering the audio signal prior to the step of causing the audio to be converted into textual data.

13. The method of claim 11 further comprising the step of buffering the audio signal until the identified user profile is transmitted to the selected one of the plurality of dictation servers.

14. The method of claim 8 further comprising the steps of:

receiving a revised textual data signal from the client station wherein the revised textual data signal comprises at least one correction of the transcription;

using the revised textual data signal to modify the identified user profile; and

storing the modified, identified user profile.

15. A system to provide a client station with a distributed dictation and transcription service comprising:

at least one dictation manager having a network connection to at least one client station, the network connection comprising at least one of a data port to receive audio signals from the at least one client station, to transmit textual data to the at least one client station, to receive textual data from the at least one client station, and to transmit textual data to the at least one client station;

at least one dictation server comprising at least one transcription, engine to transform audio signals of dictated words to textual data signals;

at least one network connection between the at least one dictation manager and the at least one dictation server adapted to transmit audio signals and textual data signals, between the at least one dictation manager and the at least one dictation server;

the at least one dictation manager comprising a processor with a means for selecting one of the at least one dictation servers to transform audio signals of-dictated words to textual data signals; and

a memory containing at least one user profile arid information indicating which of the at least one dictation servers has been provided the at least one user profile and wherein the information is used by the means for selecting to direct the audio to a selected one of the dictation servers if the user profile has been provided and is still maintained by the selected one of the at least one dictation servers.

16. The system of claim 15 wherein the means for selecting one of the at least one dictation servers comprises a load balancer.

17. The system of claim 15 Wherein the means for selecting one of the at least one dictation server Comprises a determiniator that uses the information in memory to determine one of the at least one dictation servers that has been provided with the at least one user profile.

18. The system of claim 15 wherein the dictation manager transmits a user profile to the selected one of the at least one dictation servers.

19. The system of claim 18 wherein the audio signal received from the client station by the dictation manager is stored in a buffer while the dictation manager is transmitting the user profile.

20. The system of claim 15 wherein be dictation manager comprises a user profile corrector that updates the user profile based on received corrective textual data from the client station.