DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 07/07/2025 have been fully considered but they are not persuasive.
With regards to the 101 rejection the applicants amendments do overcome the rejection.
Applicants amendments required a new search and new art was found to Nelson which teaches the amended claimed limitations.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bojar U.S. PAP 2023/0306207 A1 in view of Rubin U.S. Patent No. 10,409,919 B2, further in view of Nelson U.S. PAP 2019/0108492 A1.
Regarding claim 1 Bojar teaches a method of transforming data (method of real time speech translation, see abstract), comprising:
receiving a first text data set, wherein the first text data set comprises alphanumeric information that comprises a plurality of words in a first language (Transcribing, in real time, at least part of the input data and interpretation data into text data using at least one automatic speech recognition (ASR) system, see par. [0011]);
generating a second text data set, wherein the second text data set comprises the first text data set translated from the first language to a second language (Translating, in real time, the text data into one common language using at least one machine translation (MT) system, see par. [0012]);
generating a third text data set, wherein the third text data set is formed by translating the second text data set from the second language to the first language (It is also possible to translate into a language that is the source language or one of the translation languages, e.g., in order to provide a transcript of the speech or human interpretation which might be of higher quality then the one provided by ASR in the ASR step, see figure 1 and par. [0058]).
However Bojar does not teach simultaneously displaying, by use of a first program running on a first computer, the second text data set and the third text data set.
In the same field of endeavor Rubin teaches in FIG. 10 an example of screens displaying the translated and translate-backed texts on a display area on the display device 10. The upper screen shows three kinds of texts: an input text “Watch for this” input by user A; translated texts “Ver para este” (Spanish) and “Échanger ma montre” (French); and translate-backed texts “See this” and “Sell my clock.”, see col. 7 lines 9-26. See also claim 3.
It would have been obvious to one of ordinary skill in the art to combine the Bojar invention with the teachings of Rubin for the benefit of promoting communication of users between different languages, see background col. 1 lines 29-34.
However Bojar in view of Rubin do not teach determining that one or more words in the plurality of words in the first text data have a quality score least than a threshold quality score; and wherein the third text data further comprises the one or more words of the plurality of words of the first text data having the quality score less than the threshold score and the one or more words of the third data set comprise an embellishment text based on the quality score.
In the same field of endeavor Nelson teaches that some units of speech for which all of the confidence scores are below a specified threshold may be designated for supplemental processing, see par. [0298]. As depicted in FIG. 14, the confidence scores may for each translated/transcribed unit of speech may be included in resulting translation/transcription data 1162 and may be used during subsequent processing. For example, confidence scores may be displayed on a graphical user interface concurrent with the display of resulting translation/transcription data 1162 to provide a visual indication of the accuracy of translated/transcribed units of speech. Confidence scores may also be included in various types of documentation, such as a meeting summary or meeting transcript. Special visual effects, such as highlighting, formatting, etc., may be used to conspicuously identify particular units of speech that have a confidence score below a specified threshold, to trigger, for example, manual editing or special processing, see par. [0300].
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin invention with the teachings of Nelson for the benefit of allowing for manual or special processing of low confidence speech, see par. [0300].
Claim(s) 3-4, 7-16, 19 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bojar U.S. PAP 2023/0306207 A1, in view of Rubin U.S. Patent No. 10,409,919 B2, in view of Nelson U.S. PAP 2019/0108492 A1 further in view of Lemon U.S. Patent No. 10,431,216 B1.
Regarding claim 2 Bojar in view of Rubin in view of Nelson does not teach the method of claim 1, further comprising: generating a fourth text data set that comprises the first text data set, wherein generating the fourth text data set comprises determining that one or more words within the first text data set are to be illustrated as embellished text when displayed; and simultaneously displaying, by use of the first program, the second text data set, the third text data set, and the fourth text data set.
In the same field of endeavor Lemon teaches systems and methods that provide a user experience for transcribing audio and/or video messages and providing mobile devices and other types of devices with enhanced transcriptions. The enhanced transcriptions may include visual emphasis indicators (embellishments), language translations, and video message transcriptions. By so doing, the present systems and methods allow users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20.
Lemon teaches wherein generating the fourth text data set comprises determining that one or more words within the first text data set are to be illustrated as embellished text when displayed (transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56);
and simultaneously displaying, by use of the first program, the second text data set, the third text data set, and the fourth text data set (Sending the text data representing the transcription to the second user interface 140 may cause the second device 122 to display the transcription, or a portion thereof, on the second user interface 140. The transcription may be displayed as typed text and may include the emphasized words and/or punctuation as described herein, see col. 6 lines 2-10).
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin in view of Nelson invention with the teachings of Lemon for the benefit of allowing users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20.
Regarding claim 3 Lemon teaches the method of claim 2, wherein the embellished text comprises a different visual characteristic from the other words within the second text data set, and the different visual characteristic comprises at least one of bolding, italicizing, underlining, highlighting, or blurring (the transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
Regarding claim 4 Lemon teaches the method of claim 2, further comprising:
altering, by use of the first program, at least one of the embellished text in the fourth text data set to form a fifth text data set (editing of a transcription, see col. 27, lines 12-13);
generating sixth text data set, wherein the sixth text data set comprises the fifth text data set translated from the first language to a second language (The application and/or the remote system may translate the transcription and provide text representing a translated transcription to the recipient, see col. 2, lines 60-62);
Bojar teaches generating seventh text data set, wherein the seventh text data set is formed by translating the sixth text data set from the second language to the first language (It is also possible to translate into a language that is the source language or one of the translation languages, e.g., in order to provide a transcript of the speech or human interpretation which might be of higher quality then the one provided by ASR in the ASR step, see figure 1 and par. [0058]);
and simultaneously displaying, by use of a first program running on a first computer, the fifth text data set, the sixth text data set, and the seventh text data set (The final translations are preferably provided to the end users in real time, e.g., they are sent to their user devices via a web client as soon as given MT system considers a part of translation to be final. , see par. [0058]; The output from the currently most credible source is then sent to another MT system(s), e.g., to three different systems for translation from English into Italian, Spanish and Swedish. Three texts in these three final translation languages are therefore provided from the method., see par. [0082]).
Regarding claim 7 Bojar teaches the method of claim 2, wherein the first text data set is generated from a transcription of audible information received from a user (receiving, in real time, input data comprising a source speech in a source language, transcribing at least part of the input data into text data using ASR system, see par. [0009-0011]).
Regarding claim 8 Lemon teaches the method of claim 7, wherein the embellished text comprises a first embellished term and a second embellished term, wherein the first embellished term and the second embellished term each comprise a different characteristic, and wherein the different characteristic comprises at least one of bolding, italicizing, underlining, highlighting, or blurring of the embellished text (the transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
Regarding claim 9 Lemon teaches the method of claim 8, wherein the different characteristic between the first embellished term and the second embellished term is based on metadata that is associated with the first text data set (transcription 1510 may be displayed different to visually indicate that the audio corresponding to the audio data is being output. For example, all or a portion of the transcription 1510 may change color, change font style, change font size, be highlighted, be underlined, be italicized, and/or be bolded, see col. 25, lines 56-61).
Regarding claim 10 Lemon teaches the method of claim 9, wherein the metadata further comprises: first metadata associated with the first embellished term (transcription 1510 may be displayed different to visually indicate that the audio corresponding to the audio data is being output. For example, all or a portion of the transcription 1510 may change color, change font style, change font size, be highlighted, be underlined, be italicized, and/or be bolded, see col. 25, lines 56-61); and second metadata that is associated with the second embellished term, wherein the first metadata and second metadata include information relating to contextual information provided in the audible information (he name 1414 may also include additional identifying information about the second user, such as, for example, whether the conversation is with a device classified as a home device or a work device, and/or whether the conversation is with multiple devices located in the same environment or associated with a group of users. For example, the additional identifying information may be that the conversation is with a group of users, see col. 24, lines 3-20).
Regarding claim 11 Lemon teaches the method of claim 1, further comprising: opening an application on the first computer; and receiving a triggering event that causes the first program to open prior to receiving the first text data set (The instruction may also correspond to the first user 130 speaking or otherwise entering a command to send the audio data and/or to start a conversation with second user 132 and/or the second user's profile and/or account, see col. 6, lines 45-49).
Regarding claim 12 Lemon teaches the method of claim 11, further comprising inserting the second text data set into the opened application (The second text data representing the second transcription may be generated in a similar manner to the transcription of the audio data described herein. The remote system 126 may send the second audio data and the second text data representing the second transcription to the first user interface 136 and to the second user interface 140, see col. 7 lines 6-9).
Regarding claim 13 Bojar teaches a method of transforming data, comprising:
receiving, by use of the first program, a first text data set, wherein the first text data set comprises alphanumeric information that comprises a plurality of words in a first language(Transcribing, in real time, at least part of the input data and interpretation data into text data using at least one automatic speech recognition (ASR) system, see par. [0011]);
and simultaneously displaying, by use of the first program, the first text data set and the second text data set (The final translations are preferably provided to the end users in real time, e.g., they are sent to their user devices via a web client as soon as given MT system considers a part of translation to be final. , see par. [0058]; The output from the currently most credible source is then sent to another MT system(s), e.g., to three different systems for translation from English into Italian, Spanish and Swedish. Three texts in these three final translation languages are therefore provided from the method., see par. [0082]).
Nelson teaches Determining that two or more words of the plurality of words in the first text data have a quality score less than a threshold quality score (units of speech for which all of the confidence scores are below a specified threshold may be designated for supplemental processing, as described in more detail hereinafter, see par. [0298]).
However Bojar in view of Rubin in view of Nelson does not teach opening an application on a first computer; receiving a triggering event that causes a first program, running on the first computer, to open; and reformatting the first text data set to form a second text data set, wherein reformatting the first text data set comprises adjusting a spatial relationship between or visual characteristic of the at least two of the plurality of words in the first text data have a quality score less than a threshold quality score.
In the same field of endeavor Lemon teaches systems and methods that provide a user experience for transcribing audio and/or video messages and providing mobile devices and other types of devices with enhanced transcriptions. The enhanced transcriptions may include visual emphasis indicators (embellishments), language translations, and video message transcriptions. By so doing, the present systems and methods allow users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20.
Lemon teaches receiving a triggering event that causes a first program, running on the first computer, to open (The instruction may also correspond to the first user 130 speaking or otherwise entering a command to send the audio data and/or to start a conversation with second user 132 and/or the second user's profile and/or account, see col. 6, lines 45-49); and reformatting the first text data set to form a second text data set, wherein reformatting the first text data set comprises adjusting a spatial relationship between or visual characteristic of at least two of the plurality of words (transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin in view of Nelson invention with the teachings of Lemon for the benefit of allowing users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20.
Regarding claim 14 Lemon teaches the method of claim 13, wherein the adjusting the visual characteristic of two or more words of the plurality of words comprises bolding, italicizing, underlining, highlighting, or blurring the two or more words(transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
Regarding claim 15 Lemon teaches the method of claim 14, further comprising: forming a third text data set by altering, by use of the first program, one of the plurality of words in the first text data set (editing of a transcription, see col. 27, lines 12-13);
reformatting the third text data set to form a fourth text data set, wherein reformatting the third text data set comprises adjusting a spatial relationship between or visual characteristic of at least two of the plurality of words within the third text data set (The text of the text message and/or transcription may be presented in a text input window 1908 along with a keyboard 1910. The user may utilize the keyboard 1910 to edit the text message and/or transcription. The edited text message and/or transcription may be presented in the text input window 1908 while editing is in progress, see col. 27 lines 12-20);
Bojar teaches and simultaneously displaying, by use of the first program, the third text data set and the fourth text data set (The final translations are preferably provided to the end users in real time, e.g., they are sent to their user devices via a web client as soon as given MT system considers a part of translation to be final. , see par. [0058]; The output from the currently most credible source is then sent to another MT system(s), e.g., to three different systems for translation from English into Italian, Spanish and Swedish. Three texts in these three final translation languages are therefore provided from the method., see par. [0082]).
Regarding claim 16 Bojar teaches the method of claim 14, wherein the first text data set is generated from a transcription of audible information received from a user (receiving, in real time, input data comprising a source speech in a source language, transcribing at least part of the input data into text data using ASR system, see par. [0009-0011])..
Regarding claim 19 Lemon teaches the method of claim 18, wherein the different characteristic between the first embellished term and the second embellished term is based on metadata that is associated with the first text data set (transcription 1510 may be displayed different to visually indicate that the audio corresponding to the audio data is being output. For example, all or a portion of the transcription 1510 may change color, change font style, change font size, be highlighted, be underlined, be italicized, and/or be bolded, see col. 25, lines 56-61).
Regarding claim 20 Bojar teaches a method of transcribing audible information comprising:
receiving, by a program implemented on a computer, first text data set, wherein the first text data set comprises alphanumeric information that comprises a plurality of words in (Transcribing, in real time, at least part of the input data and interpretation data into text data using at least one automatic speech recognition (ASR) system, see par. [0011]);
generating, by the program, second text data, wherein the second text data comprises the plurality of words of the first text data set formatted based on a format for an application the second text data set is to be transmitted on (Translating, in real time, the text data into one common language using at least one machine translation (MT) system, see par. [0012]).
However Bojar in view of Rubin does not teach wherein a special relationship between the plurality of words in the second text data is different from a spatial relationship of the plurality of words in the first texted data; transmitting, by the program, the second text data set to the application; and displaying by use of the program running on the computer the first text data set and the second text data set.
Nelson teaches displaying by use of the program running on the computer the first text data set and the second text data set (confidence scores may be displayed on a graphical user interface concurrent with the display of resulting translation/transcription data 1162 to provide a visual indication of the accuracy of translated/transcribed units of speech. Confidence scores may also be included in various types of documentation, such as a meeting summary or meeting transcript. Special visual effects, such as highlighting, formatting, etc., may be used to conspicuously identify particular units of speech that have a confidence score below a specified threshold, see par. [0300]).
In the same field of endeavor Lemon teaches systems and methods that provide a user experience for transcribing audio and/or video messages and providing mobile devices and other types of devices with enhanced transcriptions. The enhanced transcriptions may include visual emphasis indicators (embellishments), language translations, and video message transcriptions. By so doing, the present systems and methods allow users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20. Reformatting the first text data set to form a second text data set, wherein reformatting the first text data set comprises adjusting a spatial relationship between or visual characteristic of at least two of the plurality of words (transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
Lemon teaches that the transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56) and sending the text data representing the transcription to the second user interface 140 may cause the second device 122 to display the transcription, or a portion thereof, on the second user interface 140. The transcription may be displayed as typed text and may include the emphasized words and/or punctuation as described herein, see col. 6 lines 2-10.
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin in view of Nelson invention with the teachings of Lemon for the benefit of allowing users to send and receive transcriptions in a manner that mimics live conversation, see col. 2 lines 8-20.
Claim(s) 5, 6, 17 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bojar U.S. PAP 2023/0306207 A1 in view of Rubin U.S. Patent No. 10,409,919 B2, in view of Nelson U.S. PAP 2019/0108492 A1, in view of Lemon U.S. Patent No. 10,431,216 B1 further in view of Vanlerberghe EP 3,477,496 A1.
Regarding claim 5 Bojar teaches the method of claim 4, wherein the alphanumeric information received in the first text data set comprises the plurality of words in the first language and a quality score for each of the plurality of words (the transcribed text can also be translated in parallel by multiple MT systems, again forming multiple sources for selection. This approach might further increase the quality of the final translations, see par. [0062]).
However Bojar in view of Rubin in view of Lemon does not teach and the determining that the one or more words of the plurality of words that have a quality score less than a threshold quality score within the first text data set comprises: comparing the quality score of the one or more words with the threshold quality score; and selecting the one or more words to be embellished when the quality score less than the threshold quality score.
In the same field of endeavor Vanlerberghe teaches a revision system for revising translated texts that represent respective translations of a same source text by different translators, see abstract. comparing a quality score of the one or more words with a threshold quality score (a revision module comprising a search submodule configured to search for the translation errors in a translated text out of the set and to identify the translation errors in the translated text for acceptance or rejection by a revisor, see par. [0008]); and selecting the one or more words to be embellished when the quality score less than the threshold quality score (Marking the matching segment may be realized by background colouring the segment, e.g. with a first colour when the segment matches with an error-free segment in a previously revised translated text, with a second colour when the segment matches with an error-containing segment in a previously revised translated text, see par. [0009]).
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin in view of Lemon invention with the teachings of Vanlerberghe for the benefit of revision of a set of translated texts with improved efficiency and consistency, especially when revising a plurality of translations of the same source text by different translators, see par. [0007].
Regarding claim 6 Vanlerberghe teaches the method of claim 5, wherein the forming fourth text data further comprises: generating fifth text data set that comprises a list of related words corresponding to the one or more words in the fourth text data set having a quality score less than the threshold quality score (the revisor may identify new errors, ); and the method further comprises: altering, by use of the first program, at least one of the embellished text of the fourth text data set to form a sixth text data set, wherein altering the at least one embellished text comprises replacing the embellished text with a word from the list of related words within the fifth text data set (suggest corrections and penalties for such new errors. Upon confirmation, such new errors and possibly any corrections or penalties suggested by the revisor, will be fed back to the revision module , see par. [0033]).
Regarding claim 17 Borja teaches the method of claim 14, wherein the alphanumeric information received in the first text data set comprises the plurality of words in the first language and a quality score for each of the plurality of words(the transcribed text can also be translated in parallel by multiple MT systems, again forming multiple sources for selection. This approach might further increase the quality of the final translations, see par. [0062]).
However it does not teach the method further comprises: comparing the quality score of one or more words of the plurality of words of the first text data set with the threshold quality score; and selecting the one or more words to be embellished when the quality score less than the threshold quality score, and the simultaneously displaying the first text data set and the second text data set comprises displaying the first text data set with the one or more words being embellished.
In the same field of endeavor Vanlerberghe teaches a revision system for revising translated texts that represent respective translations of a same source text by different translators, see abstract. comparing a quality score of the one or more words with a threshold quality score (a revision module comprising a search submodule configured to search for the translation errors in a translated text out of the set and to identify the translation errors in the translated text for acceptance or rejection by a revisor, see par. [0008]); and the simultaneously displaying the first text data set and the second text data set comprises displaying the first text data set with the one or more words being embellished (Marking the matching segment may be realized by background colouring the segment, e.g. with a first colour when the segment matches with an error-free segment in a previously revised translated text, with a second colour when the segment matches with an error-containing segment in a previously revised translated text, see par. [0009]).
It would have been obvious to one of ordinary skill in the art to combine the Bojar in view of Rubin in view of Lemon invention with the teachings of Vanlerberghe for the benefit of revision of a set of translated texts with improved efficiency and consistency, especially when revising a plurality of translations of the same source text by different translators, see par. [0007].
Regarding claim 18 Lemon teaches the method of claim 17, wherein the embellished one or more words comprise a first embellished term and a second embellished term, wherein the first embellished term and the second embellished term each comprise a different visual characteristic, and wherein the different visual characteristic comprises bolding, italicizing, underlining, highlighting, or blurring the first embellished term and the second embellished term (the transcription may be displayed with the emphasized words and/or punctuation determined via speech recognition. Display of the emphasized words may be by, for example, highlighting, underlining, italicizing, bolding, changing of font size, changing of font style, and/or changing of font color, see col. 5, lines 51-56).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Pertinent prior art available on form 892.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael Ortiz-Sanchez whose telephone number is (571)270-3711. The examiner can normally be reached Monday- Friday 9AM-6PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Bhavesh Mehta can be reached on 571-272-7453. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MICHAEL ORTIZ-SANCHEZ/Primary Examiner, Art Unit 2656
Prosecution Insights