Captioned Telephone Services Improvement

Abstract

Internet Protocol captioned telephone service often utilizing Automated Speech Recognition can be utilized with conference calls to separate out each of the various parties' speech as text, such as with text bubbles differentiated by caller on a device of the user. Additionally, a prioritized vocabulary can be provided for each user that is not shared with a public so that if the user utilizes words in their speech not common in the general public, those words can be more accurately identified by the telephone service. The service may learn and apply that vocabulary and/or the user may provide words to the service.

Claims

1. A method of audio to text transcription provided by a captioner company comprising the steps of: a) a far end caller connected to a user in audio communication with a telephone number managed by a captioned telephone service manager; b) said captioned telephone service manager utilizing a captioner to transcribe audio to text of the far end caller and sending the text to a device of the user in approximately real time; and c) a third party caller separate from the far end caller and the user joining in a conference call with the user and far end caller with the captioned telephone service manager utilizing the captioner to transcribe audio to text from the third party caller and provide to the device of the user.

2. The method of claim 1 wherein text of the third party caller is differentiated from text of the far end caller for a duration of the connection as a conference call as provided to the device of the user.

3. The method of claim 2 wherein the text of the user is differentiated from text of the far end caller and the third party for the duration of the connection as a conference call as provided to the device of the user.

4. The method of claim 2 wherein differentiation of text is provided by one of different text bubble locations on the device, differing colors, and assigning identifiers to the far end caller, the third party and the user so the user can easily correlate text to one of the third party, the far end caller and the user.

5. The method of claim 4 wherein one of the far end caller calls the telephone number managed by the captioned telephone service to initiate the connection, and the user joins to the telephone number managed by the captioned telephone service after connection with the far end caller.

6. The method of claim 4 where more than one third party join the conference call with the text of the third parties are differentiated.

7. The method of claim 4 wherein the captioner is automated speech recognition software controlled by the captioned telephone service manager.

8. The method of claim 4 further comprising a prioritized vocabulary for the user which is not provided for use with a similar priority to a public outside of the user.

9. The method of claim 8 wherein the user provides words to the captioned telephone service manager which may be less common than used by the public.

10. The method of claim 9 wherein the words are one of geographic and technical in nature.

11. The method of claim 9 wherein the user ascribes a likely priority to words likely to be used in conversations with the user.

12. The method of claim 8 wherein the captioned telephone service manager provides a prioritized vocabulary for each user based on communications of the user.

13. The method of claim 1 wherein one of the far end caller calls the telephone number managed by the captioned telephone service to initiate the connection, and the user joins to the telephone number managed by the captioned telephone service after connection with the far end caller.

14. The method of claim 1 wherein one of the far end caller calls the telephone number managed by the captioned telephone service to initiate the connection, and the user joins to the telephone number managed by the captioned telephone service after connection with the far end caller.

15. The method of claim 4 wherein the captioner is automated speech recognition software controlled by the captioned telephone service manager.

16. A method of audio to text transcription provided by a captioner company comprising the steps of: a) a far end caller connected to a user in audio communication with a telephone number managed by a captioned telephone service manager; b) said captioned telephone service manager utilizing a captioner to transcribe audio to text of the far end caller and sending the text to a device of the user in approximately real time; and c) the captioned telephone service manager maintains a prioritized vocabulary for the user not shared with a larger public.

17. The method of claim 16 further comprising the step of: a third party caller separate from the far end caller and the user joining in a conference call with the user and far end caller with the captioned telephone service manager utilizing the captioner to transcribe audio to text from the third party caller and provide to the device of the user.

18. The method of claim 16 wherein the captioned telephone service manager provides a prioritized vocabulary for each user based on communications of the user.

19. The method of claim 16 wherein the user provides words to the captioned telephone service manager which may be less common than used by the public.

20. The method of claim 19 wherein the user ascribes a likely priority to words likely to be used in conversations with the user.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings illustrate preferred embodiments of the invention and, together with the description, serve to explain the invention. The drawings may not show elements to scale. These drawings are offered by way of illustration and not by way of limitation:

[0022] FIG. 1 is a schematic view showing two devices not associated with an IPCTS system connected to a number of an IPCTS system to provide IPCTS text to a user.

[0023] FIG. 2 is a schematic view showing a conference all providing separated text from each caller by the IPCTS system.

[0024] FIG. 3 is a schematic view of a personalized search engine utilized with some embodiments of the present invention;

[0025] FIGS. 4, 5, and 6 are flowcharts of exemplary methods for providing captioned telephone service (CTS) text for the embodiments of FIGS. 1-3, respectively, of currently preferred embodiments of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] FIG. 1 shows a presently preferred embodiment of the present invention of two people talking from phones 12,14 that are not connected to an IPCTS service 20. This could happen if the user has an old phone line that is not connected to the service or if an individual such as a doctor has an old phone number of the user (not connected to the IPCTS service) that is called by that individual. The user may conference in or otherwise connect to the IPCTS service number which is shown with a phone 16 in phantom. Phone 16 may then actually be phone 14, and allow the user to communicate back and forth with the IPCTS service 20 to receive the text bubbles 22,24 almost in real time as it exchanges between the users of phones 12,14. Take in mind that the text 22,24 is expected in many embodiments to appear on screen of the user's phone 14 during this process for many embodiments.

[0027] A flow chart of this embodiment is shown in FIG. 4 of this process 100 starting at start point 102 with the call initiated between two non-IPCTS numbers at step 104. Then an IPCTS number is conferenced in or otherwise joined at step 106 which belongs to one of the two parties 12,14. Then the text will display on the IPCTS device 108 which could very well be the device 14 or otherwise. Once the call is over the process 100 ends at step 110.

[0028] FIG. 2 shows an embodiment with multiple devices 32,34,36,38 connected in conference call such as by calling to a service 40 or it being hosted by any one of the various devices 32,34,36,38, etc. There could be additional devices as well for various embodiments. Device 38 is shown as being an IPCTS subscriber communicating back and forth with service 42 so that each party's text bubbles 44,46,47,48 appears in as segregated by caller (from devices 32,34,36,38, respectively) in one of various ways as understood by those of ordinary skill in the art. Text bubbles 44,46,47,48 could be differentiated by color, they could be differentiated by placement on the screen and/or other effects such as understood by those of ordinary skill in the art. However, no one is known in the art to provide IPCTS conference calling capability whereby the text bubbles each of the speakers are separated identified by caller. This service could also be combined with the embodiment of FIG. 1 and FIG. 4 whereby during a conference call the IPCTS subscriber (initially using a non-IPCTS service) could connect to an IPCTS number of service 20,42 which could then identify separate callers devices 32,34,36,38 possibly by the characteristics of their voice and still assign text bubbles 44,46,47,48 to each of the specific callers for various embodiments. Other technology could be utilized to identify the various speakers as well.

[0029] FIG. 5 shows an embodiment of a flowchart 200 beginning at start 202 and then initiating conference call at step 204 with IPCTS user. The system 42 transcribes the text while identifying who said what in step 206 as text bubbles 44,46,47,48 or otherwise before the call ends at step 208. The transcribed text bubbles 44,46,47,48 are shown appearing on the device 38 of user as is shown in FIG. 2.

[0030] Finally, FIG. 3 shows an embodiment where an IPCTS user device 50 conducting a call with another party 52 whereby the IPCTS system 54 transcribes text between the two parties but since user 50 believes they utilize certain words more often than the general public, they provide specific words to the IPCTS system 54 to be utilized with a personal database at 56 which could include such difficult words to spell as various medical terms, location terms, or other words that the user 50 believes would normally be difficult for the system 54 to recognize. Additionally, these specific words on the database 56 may not be common to the generally public and do not need to be utilized in an automatic speech recognition software to add to the general public due to their relatively low frequency of use amongst the general public.

[0031] Accordingly, the database 56 is not passed to other users as normally happen in ASR technology. One likely word is “deaf” which is utilized with great frequency as compared to the general public as often the pronunciation of this words is confused with “death” which might otherwise appear more frequently in the text streams as a miscommunication of the term deaf by certain ASR software. Accordingly, the user can help assist the algorithm identify which words are more frequently utilized for themselves than the general public. How much weight is given to any particular word could be at least partially controlled by the user. Furthermore, these preferences could change over time, daily or at other periods. Priorities could range from above average, to high priority, etc.

[0032] FIG. 6 shows a process 300 starting at step 302 with the user providing personalized database of words which could be updated (at any time) at step 304 and transcribing text appears at step 306 with the user be given an opportunity to update the database in step 308 before the process ends at step 310. The database 56 or algorithm of preferential selection and/or prioritization of words utilized by that specific user may be utilized by the service 50 without contribution to a larger population.

[0033] Numerous alterations of the structure herein disclosed will suggest themselves to those skilled in the art. However, it is to be understood that the present disclosure relates to the preferred embodiment of the invention which is for purposes of illustration only and not to be construed as a limitation of the invention. All such modifications which do not depart from the spirit of the invention are intended to be included within the scope of the appended claims.