Systems and methods for improvements to user experience testing
Systems and methods for transcription analysis of a recording are provided. The recording includes an audio and screenshot/video portion. The audio portion is transcribed using a machine learned model. Models may be selected by the recording quality and potentially accents or other speech patterns that are present. The transcription is then linked to the video/screen capture chronology, so that automatic scrolling is enabled, clip selection from the transcription, and searching to a video time is possible. There is improvements to user experience question generation, review of study results, and in managing the study participants.
1. A method for advanced analysis of a recording comprising:
receiving a recording comprising an audio portion, a screen capture recording and a camera video recording collected by inserting a virtual tracking code into a target web site;
processing, on a dedicated analytics server, the audio and video recording by a neural network to detect emotions of a participant;
identifying key emotions from the detected emotions; and
receiving at the server one of the key emotions on a graphical user interface by a researcher from a third party computer system responsive to the key emotions and automatically generating, at the server, a video clip, which is a subset of the video portion of the recording, wherein the start time of the video clip is based upon determining a length of time before the key emotion as the shorter of a threshold time and a delay time between an end of a word directly before a first word before the key emotion, and the end time of the video clip is based upon the selection, and wherein the selection is synchronized to the video clip.
2. The method of claim 1 , wherein the recording is audio and video recording of a participant engaged in a user experience study.
3. The method of claim 2 , further comprising processing the video recording by a neural network.
4. The method of claim 1 , wherein the processing further includes eye tracking.
5. The method of claim 1 , further comprising receiving a selection of a transcription and automatically generating a second video clip of the recording based upon the selection, wherein the selection is synchronized to the video clip.
6. The method of claim 5 , further comprising searching the transcription by a keyword.
7. The method of claim 1 , further comprising appending an annotation to a flag added to the recording.
8. The method of claim 7 , wherein the appended annotation is searchable by the keyword.
9. The method of claim 1 , further comprising aggregating a plurality of recordings.
10. The method of claim 9 , further comprising filtering the plurality of recordings by success criteria, participant attribute, and keywords.
11. The method of claim 1 , wherein the key emotions includes frustration.
12. The method of claim 11 , further comprising isolating a segment of the audio and video recordings when a success criterion has been met and frustration is detected.
13. A computerized system comprised of servers including processors and computer memory for advanced analysis of a recording comprising:
an interface for receiving a recording comprising an audio portion and at least one of a screen capture recording and a camera video recording collected by inserting a virtual tracking code into a target web site;
and
a dedicated analytics server, comprising a processor and memory, for processing the audio and video recording by a neural network to detect emotions of a participant, identifying key emotions from the detected emotions; and
a clip server, comprising a processor and memory, for generating a video clip, wherein the clip server receives one of the key emotions on a graphical user interface by a researcher from a third party computer system responsive to the key emotions, wherein the video clip is a subset of the video portion of the recording, wherein the start time of the video clip is based upon determining a length of time before the key emotion as the shorter of a threshold time and a delay time between an end of a word directly before a first word before the key emotion, and the end time of the video clip is based upon the selection, and wherein the selection is synchronized to the video clip.
14. The system of claim 13 , wherein the recording is audio and video recording of a participant engaged in a user experience study.
15. The system of claim 14 , further comprising a neural network for processing the video recording.
16. The system of claim 13 , wherein the processing further includes tracking.
17. The system of claim 13 , further comprising a video editor processor for receiving a selection of a transcription and automatically generating a second video clip of the recording based upon the selection, wherein the selection is synchronized to the video clip.
18. The system of claim 17 , further comprising an analytics processor for searching the transcription by a keyword.
19. The system of claim 13 , wherein the dedicated analytics server is further configured to append an annotation to a flag added to the recording.
20. The system of claim 19 , wherein the appended annotation is searchable by the keyword.
21. The system of claim 13 , further comprising a database for aggregating a plurality of recordings.
22. The system of claim 21 , further comprising an analytics processor for filtering the plurality of recordings by success criteria, participant attribute, and keywords.