Techniques to facilitate personalized, text messaging campaigns at scale
Described herein are techniques for integrating personalized video clips, and video-based stories with a text messaging platform or service. In some instances, an administrative end-user establishes a text messaging campaign, and designates one or more agent end-users to assist in executing the campaign. The admin end-user may designate specify one or more quality assurance routines that are to be executed against video clips recorded by agent end-users, to ensure the video clips satisfy various quality assurance requirements.
1 . A computer-implemented method comprising:
at a text messaging service, receiving via a first user interface presented to a first end-user a templated video script having at least one reference to a variable;
causing a user interface to be presented to a second end-user, the user interface including a video script derived from the templated video script and prompting the second end-user to record a video clip with a spoken message based on a message conveyed by the video script;
receiving at the text messaging service a video clip recorded by the second end-user;
processing the video clip in accordance with a quality assurance setting indicated by the first end-user; and
subsequent to determining that the video clip satisfies a quality assurance requirement specified by the quality assurance setting, updating a status of the video clip to indicate the video clip can be sent to a message recipient.
2 . The computer-implemented method of claim 1 , wherein the quality assurance requirement specified by the quality assurance setting requires the video clip to be reviewed and approved by the first end-user or another end-user designated by the first end-user.
3 . The computer-implemented method of claim 1 , wherein the quality assurance requirement specified by the quality assurance setting requires that the second end-user has a trust score exceeding a predetermined threshold before a video clip recorded by the second end-user can be sent to a message recipient via the text messaging service; the method further comprising:
subsequent to receiving at the text messaging service the video clip recorded by the second end-user, comparing a trust score assigned to the second end-user with the predetermined threshold; and
responsive to determining that the trust score of the second end-user exceeds the predetermined threshold, updating a status of the video clip to indicate the video clip can be sent to a message recipient.
4 . The computer-implemented method of claim 1 , wherein the quality assurance requirement specified by the quality assurance setting requires that the second end-user has a trust score exceeding a predetermined threshold before a video clip recorded by the second end-user can be sent to a message recipient via the text messaging service; the method further comprising:
subsequent to receiving at the text messaging service the video clip recorded by the second end-user, comparing a trust score assigned to the second end-user with the predetermined threshold; and
responsive to determining that the trust score of the second end-user does not exceed the predetermined threshold, adding the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
5 . The computer-implemented method of claim 1 , wherein processing the video clip in accordance with a quality assurance setting indicated by the first end-user comprises:
providing the video clip as input to a machine learned algorithm trained to identify objectionable images.
6 . The computer-implemented method of claim 1 , wherein the objectionable images comprise images involving nudity and images involving obscene hand gestures.
7 . The computer-implemented method of claim 1 , wherein processing the video clip in accordance with a quality assurance setting indicated by the first end-user comprises:
providing the video clip as input to an audio-to-text translation process that converts an audio track of the video clip to corresponding text;
comparing the corresponding text with text specified in the video script; and
responsive to determining a divergence between the corresponding text and the text specified in the video script, adding the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
8 . The computer-implemented method of claim 1 , wherein processing the video clip in accordance with a quality assurance setting indicated by the first end-user comprises:
providing the video clip as input to an audio-to-text translation process that converts an audio track of the video clip to corresponding text;
comparing the corresponding text with text specified in a list of known objectionable words and phrases; and
responsive to determining that text in the corresponding text includes one or more words or phrases included in the list of known objectionable words and phrases, adding the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
9 . The computer-implemented method of claim 1 , wherein processing the video clip in accordance with a quality assurance setting indicated by the first end-user comprises:
providing the video clip as input to an image processing algorithm that utilizes optical character recognition (OCR) to detect one or more words or phrases displayed in an image extracted from the video clip;
responsive to determining that a word or phrase detected in the image is included in a list of known objectionable words and phrases, adding the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
10 . A system comprising:
a memory storage device storing computer-readable instructions;
at least one processor, which, when executing the computer-readable instructions, causes the system to:
at a text messaging service, receive via a first user interface presented to a first end-user a templated video script having at least one reference to a variable;
cause a user interface to be presented to a second end-user, the user interface including a video script derived from the templated video script and prompting the second end-user to record a video clip with a spoken message based on a message conveyed by the video script;
receive at the text messaging service a video clip recorded by the second end-user;
process the video clip in accordance with a quality assurance setting indicated by the first end-user; and
subsequent to determining that the video clip satisfies a quality assurance requirement specified by the quality assurance setting, update a status of the video clip to indicate the video clip can be sent to a message recipient.
11 . The system of claim 10 , wherein the quality assurance requirement specified by the quality assurance setting requires the video clip to be reviewed and approved by the first end-user or another end-user designated by the first end-user.
12 . The system of claim 10 , wherein the quality assurance requirement specified by the quality assurance setting requires that the second end-user has a trust score exceeding a predetermined threshold before a video clip recorded by the second end-user can be sent to a message recipient via the text messaging service, the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
subsequent to receiving at the text messaging service the video clip recorded by the second end-user, compare a trust score assigned to the second end-user with the predetermined threshold; and
responsive to determining that the trust score of the second end-user exceeds the predetermined threshold, update a status of the video clip to indicate the video clip can be sent to a message recipient.
13 . The system of claim 10 , wherein the quality assurance requirement specified by the quality assurance setting requires that the second end-user has a trust score exceeding a predetermined threshold before a video clip recorded by the second end-user can be sent to a message recipient via the text messaging service, the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
subsequent to receiving at the text messaging service the video clip recorded by the second end-user, compare a trust score assigned to the second end-user with the predetermined threshold; and
responsive to determining that the trust score of the second end-user does not exceed the predetermined threshold, add the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
14 . The system of claim 1 , the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
provide the video clip as input to a machine learned algorithm trained to identify objectionable images.
15 . The system of claim 10 , wherein the objectionable images comprise images portraying nudity and images portraying obscene hand gestures.
16 . The system of claim 10 , the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
provide the video clip as input to an audio-to-text translation process that converts an audio track of the video clip to corresponding text;
compare the corresponding text with text specified in the video script; and
responsive to determining a divergence between the corresponding text and the text specified in the video script, add the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
17 . The system of claim 10 , the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
provide the video clip as input to an audio-to-text translation process that converts an audio track of the video clip to corresponding text;
compare the corresponding text with text specified in a list of known objectionable words and phrases; and
responsive to determining that text in the corresponding text includes one or more words or phrases included in the list of known objectionable words and phrases, add the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.
18 . The system of claim 10 , the memory storage device storing additional computer-readable instructions, which, when executed by the processor, cause the system to:
provide the video clip as input to an image processing algorithm that utilizes optical character recognition (OCR) to detect one or more words or phrases displayed in an image extracted from the video clip;
responsive to determining that a word or phrase detected in the image is included in a list of known objectionable words and phrases, add the video clip to a review queue for review and approval by the first end-user, or another end-user designated by the first end-user.