Abstract: A video is rendered. The video features a host and is viewed by one or more viewers. A video segment is accessed. The video segment is related to the video and is accessed by an operator. The video segment includes a performance by an individual. A synthesized video segment is created from the video segment that was accessed. The synthesized video segment includes the performance as accomplished by the host. At least one insertion point within the video is determined for the synthesized video segment. The synthesized video segment is inserted by the operator into the video at the at least one insertion point. The inserting is accomplished dynamically and appears seamless to a viewer. A remainder of the video is rendered after the at least one insertion point. The determining at least one insertion point includes a response to an interaction by the viewers of the video.

Abstract: Techniques for manipulating an expandable video loop with replacement audio are provided. A video that includes a performance by an individual is accessed. One or more processors are used to identify a minimum desired loop within the performance by the individual. Video information is extracted from the video loop and analyzed with machine learning. An audio file is obtained based on a script written by a user. The script is combined with a synthesized voice which is created from the individual video performance to obtain the audio file. The extracted video information is used to manipulate the lips of the individual based on the audio file. A second video loop is created, combining the individual from the first video loop with the video of the manipulated lips of the individual and the synthesized audio file. The second video loop is appended to the end of the first video loop.

Abstract: Disclosed embodiments provide techniques for livestream with synthetic scene insertion. A prerecorded livestream featuring a host is rendered to one or more viewers. An operator accesses a video segment related to the prerecorded livestream. The related video segment includes a performance by an individual. The operator retrieves an image of the host of the prerecorded livestream and creates from the related video segment a synthesized video segment that includes the performance of the individual accomplished by the host. One or more insertion points are determined within the prerecorded livestream for the insertion of the synthesized video segment. The operator inserts the synthesized video segment into the prerecorded livestream at the one or more determined insertion points. The insertion is accomplished dynamically and appears seamless to the viewer. The remainder of the prerecorded video is rendered to the one or more viewers after the insertion point.

Abstract: A gas destruction system using a gas destroying device communicatively coupled to an electronic controller subsystem for neutralizing unwanted gas emissions. The system includes a gas destruction device and method and can be customized to variable conditions by way of controlling available flow area. The disclosed gas destruction device includes multiple combustion manifolds separated by valves, which allow gas to flow to sets of nozzles as needed to maintain stable combustion. The system consists of multiple nozzles for gas flow, a main shutoff valve, a flow measurement sensor, and an igniter and power source. The combustion products are allowed to flow through the exhaust outlet attached to the combustor. A communicator device is included in the system to send system information, preferably wirelessly. The system can operate autonomously via local or remote controllers, or manually via local or remote commands.

Abstract: Techniques for video processing using artificial intelligence are disclosed. An interactive video interface (IVF) located in a physical store is accessed. The IVF includes a video camera, interactive screen, and one or more microphones. The IVF includes a synthetic human that interacts with customers. A customer initiates an interaction with the IVF by standing within a minimum distance of the IVF, speaking to the IVF, interacting with the IVF screen, or scanning an identification card. The IVF collects input from the customer regarding products in the store. The IVF creates responses to the customer inquiries based on a large language model (LLM), including skills such as checking product inventory, recommending related products, completing ecommerce purchases, and coordinating product deliveries. The IVF produces a video segment featuring the response performed by the synthetic human. The IVF presents the video segment to the customer, including performing the selected skills.

Abstract: Techniques for video processing using artificial intelligence are disclosed. An embedded interface included on a website and/or mobile application is accessed. The embedded interface includes one or more products for sale. A user requests an interaction based on a product for sale. A first video segment including a synthetic human is displayed to the user. The embedded interface collects user input in response to the video segment. One or more classifiers, which can comprise one or more lightweight LLMs, classify the user input, identifying a type of conversation. The user input is routed by a controller to a module which provides instructions to a final LLM. The final LLM creates a response to the interaction with the user, based on the instructions. The response is used to generate a second video segment that is displayed to the user.

Abstract: Techniques for managing artificial intelligence interactions are disclosed. A large language model (LLM) is trained, including information related to products for sale. The product information resides in a product knowledge base. Users interacting with a synthetic human generate input related to the products for sale through an embedded interface included in a website or application. The user input is captured and the LLM creates responses based on information in the product knowledge base. The responses are used to produce video segments that are presented to the user. The user creates additional input based on the video responses from the LLM. The additional input is evaluated and used to trigger self-improving steps to improve the content of the product knowledge base and the responses generated for the user. The self improving includes quality metrics; self-learning instructions; information gap identification; and information collection from third-party websites, product experts, and sellers.

Abstract: Techniques for video streaming are disclosed. Audio input related to products for sale is received from a user within an embedded interface. The embedded interface includes an artificial intelligence virtual assistant represented by a synthetic human host. The audio input is captured by a natural language processing engine, creating a data segment which is sent to a large language model (LLM). The data segment can be divided into one or more subsegments. Each subsegment can be processed independently by the LLM to generate a response based on product information stored in the LLM database. The LLM responses are converted by a text-to-speech (TTS) converter into an audio response. The audio responses are used to produce a video segment including the synthetic host performing the audio stream. The capturing and processing of user audio continues to allow a human-like dialogue with the synthetic human host.

Abstract: Techniques for video streaming are disclosed. A plurality of audio files is accessed. The audio files comprise answers from a large language model to a user requesting information on products for sale. Each audio file is used to produce a video segment, featuring an animated artificial intelligence virtual assistant performing the audio file. The animating uses a Mel spectrogram analysis of the audio file. The audio analysis allows each word, sound, or phoneme of the audio file to be aligned with the matching video frame of the virtual assistant speaking each word. As the first video segment is streamed to the user, additional audio files are used to produce more video segments. The subsequent video segments are buffered based on a probabilistic model as the previous segments are streamed, so that as each video segment is completed, the next segment begins.

Abstract: Video processing using artificial intelligence is disclosed. An embedded interface including products for sale is accessed. A user requests an interaction based on one of the products. The embedded interface initiates a video segment including a synthetic human in response to the user request. The user submits a question or comment. The interface collects the user input and converts it into a dataset readable by a large language model (LLM). The LLM generates a response to the user request. The response is used to generate an audio stream. The audio stream includes simulated human speech errors and pauses. The audio stream is segmented and a video clip is synthesized for each audio segment. The video clips are assembled into a new video segment which is presented to the user. Additional user interactions are collected and new video segments are generated in response.

Abstract: Techniques for manipulating video livestream background images are disclosed. A short-form video, such as a livestream video or livestream replay video, can be analyzed for context. Computer-implemented techniques may be used for performing entity detection, and can also detect a change in subject based on speech and/or actions of a host individual. The subject can include a particular product. The detecting a change in subject can include detecting a foreground object and identifying the foreground object as a product. The identification of the foreground object as a product can include performing optical character recognition on text imprinted on a foreground object. The identification of the foreground object as a product can include image recognition techniques. The identification of the foreground object as a product can include scanning of an optical code such as a barcode that is imprinted on the product.

Abstract: Techniques for manipulating an expandable video loop with replacement audio are provided. A video that includes a performance by an individual is accessed. One or more processors are used to identify a minimum desired loop within the performance by the individual. Video information is extracted from the video loop and analyzed with machine learning. An audio file is obtained based on a script written by a user. The script is combined with a synthesized voice which is created from the individual video performance to obtain the audio file. The extracted video information is used to manipulate the lips of the individual based on the audio file. A second video loop is created, combining the individual from the first video loop with the video of the manipulated lips of the individual and the synthesized audio file. The second video loop is appended to the end of the first video loop.

Abstract: A computer-implemented method for video analysis is disclosed. A livestream, including a livestream chat, is accessed. The livestream includes products for sale. The livestream includes timestamps based on when each product is highlighted. The timestamps allow users to jump to livestream locations based on the highlighted products. The livestream is monitored by a large language model. The large language model (LLM) detects a question from users in the livestream, and the LLM determines answers to the question. An engagement metric is calculated for each answer. The engagement metric is predictive of future engagement in the livestream by users if the answer is posted. A human assistant in the livestream reviews the LLM answer with the highest engagement score and chooses to edit the answer, to post the answer, or not to post the answer to the livestream.

Abstract: Disclosed embodiments provide techniques for video chat initiation based on machine learning. A website that includes products for sale is accessed. The website is viewed by multiple users. User information is collected and analyzed based on machine learning. The machine learning model is used to predict a purchase intention for a user regarding one or more products for sale. The prediction is based on the machine learning analysis. Information about one or more sales associates is gathered and analyzed based on machine learning. The analysis of sales associate information is used to match a user with a purchase intention with a sales associate. The sales associate can initiate a chat interaction with the user through an overlay on the website. The chat interaction can provide additional user information for the machine learning model. During the chat interaction, the user can purchase one or more products.

Abstract: Techniques for video generation based on machine learned video template usage are disclosed. A plurality of videos is obtained, and video scene analysis on each video is performed. Video cuts for each video are detected, and objects within each video are identified. The identifying includes detecting a person, face, building, or vehicle. Metadata is categorized for each of the videos based on the scene analysis, the video cuts, and the objects within the videos. Template information is stored, including the categorized metadata, on each of the videos. Each video is stored as a template video along with the template information. The template information on a subset of videos is ranked. A basis video is selected based on the template information. A further video is generated based on the basis video. The further video is stored as a further template video along with the template videos.

Abstract: Disclosed embodiments provide techniques for immediate livestreams in a short-form video ecommerce environment. A website that includes a plurality of products for sale is accessed. The website is viewed by a plurality of users. One or more groups of users is identified. Each of the one or more groups includes users with an interest in one or more products from the plurality of products for sale. An immediate livestream is created for each group of the one or more groups of users when a number of users within each group is higher than a threshold value. The immediate livestream includes the one or more products. The immediate livestream is rendered within the container unit on the website to each group of the one or more groups of users. An ecommerce purchase of the one or more products is enabled within the immediate livestream.

Abstract: Disclosed embodiments provide techniques for dynamic synthetic video chat agent replacement. A human host receives a request for a video chat initiated by a user. An image for a synthetic host including a representation of an individual is retrieved. The image of the synthetic host is selected based on information about the user. Aspects of the individual included in the image are extracted using one or more processors. A video performance by the human host responding to the statement or query by the user is captured. A synthetic host performance is created in which the video performance of the human host is dynamically replaced by the individual that was extracted so that the synthetic host performance responds to the user statement or query. The synthetic host performance is rendered to the user and supplemented with additional synthetic host performances as the video chat continues.

Abstract: Disclosed embodiments provide techniques for synthesized responses to predictive livestream questions. A livestream featuring a host is accessed and is viewed by one or more viewers. The audio, video, and images from the livestream are analyzed and a plurality of potential questions from the one or viewers is predicted based on the analysis. An answer to each potential question is generated based on a large language model (LLM) neural network. A synthesized video segment is created for each answer generated by a generative AI chatbot linked to a LLM for each potential question. During the livestream, real-time questions from the one or more viewers are detected. The real-time questions are matched to the synthesized video segment answers and the matched synthesized video segment is rendered to the one or more livestream viewers. An ecommerce environment is rendered during the livestream to allow purchasing of products.

Abstract: Disclosed embodiments provide techniques for synthesized realistic metahuman short-form videos. A photorealistic representation of an individual is accessed from media sources including videos, photographs, livestreams, or a 360-degree recording of a human host. The individual can be selected based on information on the viewer of the short-form video, such as purchase history, viewing history, and metadata. The photorealistic representation is isolated using machine learning and is used to create a three-dimensional (3D) model of the individual, based on a game engine. A realistic synthetic performance is created by combining the 3D model of the individual with animation generated by a game engine. The synthesized performance can include the voice of the human host. The synthesized performance is inserted into a metaverse environment and rendered to a viewer as a short-form video, including an ecommerce window with an on-screen product card and virtual purchase cart.

Abstract: Disclosed embodiments provide techniques for iterative AI prompt optimization for video generation. A first text template to be read by a large language model (LLM) neural network is accessed. The template includes control parameters that are populated from within a website. The populated template is submitted as a request to the LLM neural network, which generates a first video script. The first video script is used to create a first short-form video. The first short-form video is evaluated based on one or more performance metrics. The text template, short-form video, website information, and evaluation are used to train a machine learning model that is used to create a second text template. The second text template can be used to generate a second short-form video. The evaluation of iterative text templates and resulting short-form videos continues until a usable video is produced.