DETAILED ACTION
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 of this title, 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.
Claims 1, 7-8 and 14-15 are rejected under 35 U.S.C. 103 as being unpatentable over Graham (US 20240346731 A1) in view of Smullen (US 20180212904 A1).
As per claim 1, Graham teaches the claimed:
1. A method for generating customized AI model for generating video, implemented by one or more processors operatively coupled to a non-transitory computer readable storage device, on which are stored modules of instruction code that when executed cause the one or more processors to perform said method comprising the steps of:
identifying from user text of new category for generating video by analyzing context, comparing to known categories of video by using AI model to identify known or new category;
(Graham [0069]: “At 408, the processor(s) may constrain one or more of the AI models to a predefined set of prompts. … In some examples, the predefined set of prompts can be updated periodically as new prompts are discovered or to delete or otherwise modify existing ones of the predefined set of prompts.” Graham teaches the new prompts, which correspond to the user text of new category, that are discovered or modified, which incorporates comparing and analyzing as it would be required to determine whether or not the prompts are new or not. Additionally, Graham uses the AI models for the sets of prompts, which would be used for the determination of the new categories.)
generating in real time personal/customized AI model for new category by learning subject by
(Graham [0060]: “Implementing specialized models 200(A), 200(B) may improve the quality of the output data 204(A), 204(B) to tailor the AI-generated synthetic content 102 for a particular application, and/or to help achieve real-time output of synthetic content 102 (e.g., on a display)”
Graham [0014]: “the AI model(s) may be trained using sequential video frames as training data in order to generate output data … The AI model(s) can also be trained with, among other things, prompt data, which allows the trained AI model(s) to generate output data (e.g., image data) based on any suitable type of prompt(s), which, in some examples, may be provided live, by a user.”
Graham teaches the tailored AI models specific to particular applications or categories that are trained by different videos that are suitable for any type of prompt. Graham expands on this by saying in paragraph [0060]: “representing synthetic content in a particular style, such as a particular style of imagery”. This is talking about the output data of a particular type of AI model, and would have required specific types of training data to be able to produce this type of output. Thus, teaching the generation of an AI model that is trained by different types of videos and use cases, and different subjects and video structure.)
generating in real time person/customized video by applying the generated AI model of the new category using the user original prompt
(Graham [0015]: “a user can prompt the AI model(s) by describing (e.g., by speaking and/or typing words using an input device(s)) a subject (e.g., person), an object, a scene, or a combination thereof, and the prompted AI model(s) generates output data representing synthetic content (e.g., imagery), and this synthetic content is output (e.g., displayed) in real-time via an output device (e.g., a display).”
Graham [0043]: “output data 104 from each model 100 may be combined to generate video data corresponding to the video content 108 featuring the synthetic content 102 associated with each model 100. Implementing specialized models 100 may improve the quality of the output data 104 and/or help achieve real-time output of synthetic content 102”.
Graham teaches the real time video generation using the trained AI models, specifically the specialized models incorporating the new category that are trained from the user prompts.)
Graham alone does not explicitly teach the remaining claim limitations.
However, Graham in combination with Smullen teaches the claimed:
by third party AI language acting as an expert
(Smullen [0381]: “that third party artificial intelligence natural language processing modules may be interfaced into either layer. In this way, any other kind of cognitive services can be integrated into either layer. In typical embodiments, such cognitive services have an application programming interface. In such embodiments, the systems and methods of the present disclosure leverage such application programming interfaces to integrate the services.” Smullen teaches the third-party AI language module that can use the cognitive services and APIs to act as experts for a particular subject.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the third-party AI language model as taught by Smullen with the system of Graham in order to delegate tasks and obtain AI capabilities without having to fully train and maintain their own models.
As per claims 8 and 15, these claims are similar in scope to limitations recited in claim 1, and thus is rejected under the same rationale.
As per claim 7, Graham teaches the claimed:
7. The method of claim 1, wherein the generated AI models for new category are saved, enabling to retrieve upon identifying saved category in user text for generating video.
(Graham [0092]: “the programs 818 may include the photoreal synthetic content service 701 to implement the techniques and operations described herein, and the data 820 may include the various model(s) 100, 200, 300 and data 106 used to train the model(s) 100, 200, 300, as well as the media data (e.g., video data) described herein, such as video data corresponding to synthetic content and/or the video content, as described herein.”
Graham [0095]: “the computing device(s) 800 can have access to other computer-readable storage media to store and retrieve information, such as program modules, data structures, or other data.”
Graham teaches the data that includes and stores the various AI models as well as the media data that contains the video content generated by the AI models. These models are also retrievable by the computing device, which can then be used by the processors to generate output data when identified by the prompts given by the user, which is shown throughout, for example in paragraph [0080]: “At 606, the processor(s) may generate, using a trained AI model(s) 100, 200, 300 based at least in part on the user-provided prompt data 112, 212, output data 104, 204, 304 representing synthetic content 102.”)
As per claim 14, this claim is similar in scope to limitations recited in claim 7, and thus is rejected under the same rationale.
Claims 2-3 and 9-10 are rejected under 35 U.S.C. 103 as being unpatentable over Graham in view of Smullen and in further view of Lew (US 20250166665 A1).
As per claim 2, Graham and Smullen alone do not explicitly teach the claim limitations.
However, Graham and Smullen in combination with Lew teaches the claimed:
2. The method of claim 1, wherein the personal/customized AI model is further trained based on data for different video styles including at least one of: problem, solution, advantages, advertising, humor, educational.
(Lew [0025]: “Additionally, or alternatively, a video style may correspond to a film genre (e.g., comedy, western, thriller, etc.), a type of video (e.g., music video, documentary, etc.), and/or a description of a film editing technique (e.g., rapidly changing cuts, dramatic zooming, face-focused editing, object-focused editing, etc.).” Lew teaches the different styles of video, which include comedy and documentaries, which are educational.
Lew [0057]: “Each video style may be representative of a trained model, where the trained model is trained (e.g., pre-trained to apply different aesthetic attributes to video content in accordance with a different video style.” Lew teaches the trained AI models that are based on the different video styles.)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the training of different video styles as taught by Lew with the system of Graham and Smullen in order to be able to apply different aesthetic attributes to the video content in accordance with the specific video styles.
As per claim 9, this claim is similar in scope to limitations recited in claim 2, and thus is rejected under the same rationale.
As per claim 3, Graham and Smullen alone does not explicitly teach the claimed limitations.
However, Graham and Smullen in combination with Lew teaches the claimed:
3. The method of claim 1, further creating story board images or short video or text displayed on the screen, the user can review edit or approve the story board before generating the video.
(Lew [0090]: “In some examples, the video creation interface 210 may provide real-time feedback to the user upon receiving the selection of the particular video style. For instance, the interface may display a visual confirmation, such as highlighting or animating the selected option, to inform the user that their input has been successfully registered.”
Lew [0109]: “The user interface view 400 includes a preview region positioned in the upper portion of the screen, where a preview of video content is displayed.”
Lew [0114]: “The design of user interface view 400 emphasizes usability and accessibility, providing an intuitive layout that allows users to perform complex video editing tasks with minimal effort. The placement of the preview region, timeline, and interactive controls ensures that the user can seamlessly transition between reviewing video content and applying edits.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the ability to edit and review a preview before generating the video as taught by Lew with the system of Graham and Smullen in order to allow users to easily change or perform any video editing tasks on a preview before generating the video content.
As per claim 10, this claim is similar in scope to limitations recited in claim 3, and thus is rejected under the same rationale.
Claims 4 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Graham in view of Smullen and in further view of Orozco (US 20240394945 A1).
As per claim 4, Graham and Smullen alone does not explicitly teach the claimed limitations.
However, Graham and Smullen in combination with Orozco teaches the claimed:
4. The method of claim 1 wherein the AI model is further trained to adapt its content generation to match the style and personality that best suits the target audience.
(Orozco [0020]: “In particular, the system can accurately generate digital content that matches the style of the campaign creator, relates to the theme of the campaign, identified based on the modification inputs provided to the AI model, and includes features that the target audience will engage with. By generating digital content that matches the style and theme, and, additionally, captures the target audience, the AI model improves the efficiency and accuracy of generating and serving digital content.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the AI model that suits a target audience as taught by Orozco with the system of Graham and Smullen in order to provide a more efficient and accurate AI model that generates digital content that matches a target style tailored to a target audience.
As per claim 11, this claim is similar in scope to limitations recited in claim 4, and thus is rejected under the same rationale.
Claims 5 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Graham in view of Smullen and in further view of Wang (US 12597420 B2).
As per claim 5, Graham and Smullen alone does not explicitly teach the claimed limitations.
However, Graham and Smullen in combination with Wang teaches the claimed:
5. The method of claim 1 wherein the AI model learning/training is further provided with a deep understanding of the rules and conventions governing content creation within the specific field by training the AI model of different fields which includes different industry standards, ethical guidelines, or legal constraints.
(Wang (col 21, line 48-56): “the rules engine can work alongside ML and NLP components to enhance the AI system's overall intelligence and adaptability. The rules engine can be used to: (1) Define and enforce domain-specific constraints: By incorporating expert knowledge or industry-specific guidelines into the rules engine, the AI system can adhere to specific requirements or standards, thus ensuring the AI system's output is compliant and relevant.”
Wang (col 55, line 35-42): “Furthermore, determining if the most relevant intent and objective predicted by AI are harmful requires incorporating ethical considerations into AI systems. This can involve developing guidelines and frameworks for ethical AI, such as those focused on fairness, transparency, and accountability. Additionally, incorporating human experts into the development and testing of AI systems can provide valuable insights into potential harms and unintended consequences.”
Wang (col 30, line 58-60 – col 31, line 37-42): “The AI system also monitors and logs all user activities involving sensitive data to detect and prevent unauthorized access or misuse 803. … Legal information: Data related to legal proceedings, disputes, or investigations, such as court records, contracts, or attorney-client communications, may be deemed sensitive due to its potential impact on reputations, relationships, or legal outcomes.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the AI model that adheres to the rules of content creation as taught by Wang with the system of Graham and Smullen in order to prevent misuse of sensitive information, focus on fairness and accountability, and follow the laws established surrounding AI.
As per claim 12, this claim is similar in scope to limitations recited in claim 5, and thus is rejected under the same rationale.
Claims 6 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Graham in view of Smullen and in further view of Christensen (US 20240414394 A1).
As per claim 6, Graham and Smullen alone does not explicitly teach the claimed limitations.
However, Graham and Smullen in combination with Christensen teaches the claimed:
6. The method of claim 1 further comprising the step of: data collection and preprocessing including
Gather a vast collection of videos across various categories.
Categorize videos by themes/styles including: problem solving advertising, humor, education.
(Christensen [0024]: “Media source 140 may organize videos by one or more categories … For example, media source 140 may enable a user to find videos that are algebra tutorials.” Christensen teaches the media source that organizes the videos into one or more categories, this would also incorporate gathering the collection of videos at some point to be able to organize the videos. Thus, teaching the gathering and organizing of the videos across different categories, including educational.)
Use a large language model to transcribe video content and store metadata.
(Christensen [0098]: “Analysis system 102 may process the extracted speech using one or more techniques such as ASR to transcribe the text from the speech and generate text data 402.” Christensen teaches the analysis system that is able to transcribe text from speech and store, as stated in paragraph [0054]: “Memory 268 may store data provided by one or more components of analysis system 202.”. Additionally, Christensen uses a language model to transcribe the video in paragraph [0100]: “Text encoder 406 may use GPT2 as taking an autoregressive approach to language modeling”)
Extract features related to the structure of videos.
Christensen [0006]: “an analysis system may generate and/or extract tokens from a video and process the tokens to create features of the video which are associated with one or more labels in a multi-label classification.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the transcription of videos as taught by Christensen with the system of Graham and Smullen in order to allow models to process and understand language more naturally and to interpret and respond more accurately.
As per claim 13, this claim is similar in scope to limitations recited in claim 6, and thus is rejected under the same rationale.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA SUO whose telephone number is (571) 272-8387. The examiner can normally be reached Mon-Fri 8am-5pm.
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, Daniel Hajnik can be reached on (571) 272-7642. 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.
/JOSHUA SUO/Examiner, Art Unit 2616
/DANIEL F HAJNIK/Supervisory Patent Examiner, Art Unit 2616
Prosecution Insights