My stuff

  • My Workflows

  • Liked Workflows

  • Following Workflows

Go to OpenArt main site
Upload workflow

Temporal Prompt Engine for ComfyUI-CogVideoX. Coherent(ish) Video + Audio Pipeline

5.0

0 reviews
4
6.6K
642
4
Description


Check out our Temporal Prompt Generator for Windows and Linux

 https://github.com/TemporalLabsLLC-SOL/TemporalPromptGenerator 



This is completely capable of longer than 6 second generations.

The video above is at 30 steps.

Be Ambitious.

# Temporal Prompt Engine: Intuitive Scene Builder for Video and Audio Generation


---


### Table of Contents

1. [Introduction](#1-introduction)

2. [Features Overview](#2-features-overview)

3. [Installation](#3-installation)

  - [Prerequisites](#prerequisites)

  - [Automated Installation](#automated-installation)

  - [Manual Installation](#manual-installation)

    - [Windows](#windows)

    - [Linux](#linux)

4. [Quick Start Guide](#4-quick-start-guide)

5. [API Key Setup](#5-api-key-setup)

6. [Harnessing the Power of ComfyUI](#6-harnessing-the-power-of-comfyui)

7. [Join the Temporal Labs Journey](#7-join-the-temporal-labs-journey)

8. [Donations and Support](#8-donations-and-support)

9. [Contact](#9-contact)

10. [Acknowledgments](#10-acknowledgments)


---


<a name="1-introduction"></a>


## 1. Introduction


Welcome to the **Temporal Prompt Engine**, your ultimate tool for crafting immersive video and audio experiences. This engine empowers you to generate high-quality prompts with unparalleled control over cinematic elements, all while being intuitive and accessible for both Windows and Linux users.


### **Unleash Your Creativity**


Imagine directing a scene set in a cyberpunk future, with neon-lit streets captured through the lens of a 1980s camera, and accompanied by a rich, layered soundscape that adapts dynamically to your visuals. With the **Temporal Prompt Manager**, this isn't just possible—it's straightforward and exhilarating.


The **Temporal Prompt Manager** allows you to select standard film terms through user-friendly dropdown menus combined with a basic input concept. These selections are transformed into detailed lists of visual and audio prompts, ensuring that every aspect of your scene aligns perfectly with your creative vision. Additionally, the engine is capable of generating custom sound effects using the powerful **AudioLDM2**, giving you the tools to create truly unique and immersive audio landscapes.


### **Why Temporal Prompt Engine?**


- **Intuitive Interface**: Designed for both beginners and professionals, making complex scene creation accessible to everyone.

- **Advanced Customization**: Select from a wide range of film terms to fine-tune your prompts.

- **Seamless Integration**: Works flawlessly with tools like ComfyUI for enhanced video generation.

- **Cross-Platform Compatibility**: Available for both **Windows** and **Linux** systems, ensuring broad accessibility.

- **Community-Driven**: Join a vibrant community of innovators, developers, and artists pushing the boundaries of AI-driven media creation.


Whether you're a filmmaker, game developer, digital artist, or an AI enthusiast, the Temporal Prompt Engine is designed to unleash your creativity and bring your visions to life with precision and style.


---


<a name="2-features-overview"></a>


## 2. Features Overview


### **Unleash Your Creativity with Advanced Controls**


- **Cinematic Video Prompts**: Tailor every aspect of your scene—from camera type and lens to lighting and framing.

- **Adaptive Audio Prompts**: Generate immersive soundscapes that perfectly match your visuals.

- **Historical and Futuristic Cameras**: Choose from an extensive list of camera models across decades, adding authenticity or a futuristic touch to your scenes.

- **Dynamic Variables**: Adjust settings like resolution, frame rate, and aspect ratio to guide the AI in generating content that matches your vision.

- **Special Modes**: Incorporate thematic elements with Holiday Mode or add creative unpredictability with Chaos Mode.

- **Interconnected Settings**: Experience how choices like selecting a vintage camera influence other variables like resolution and aspect ratio, creating a cohesive and authentic output.


### **Intuitive and User-Friendly Interface**


- Designed for simplicity and ease of use, allowing both beginners and professionals to navigate effortlessly.

- Compatible with both **Windows** and **Linux** systems.


---


<a name="3-installation"></a>


## 3. Installation


Setting up the Temporal Prompt Engine is simple and hassle-free, allowing you to focus on creation rather than configuration. This section provides both automated and manual installation steps for **Windows** and **Linux** users.


### <a name="prerequisites"></a>Prerequisites


Before installing the Temporal Prompt Engine, ensure that your system meets the following requirements:


- **Operating System**: Windows 10 or later / Linux (Ubuntu, Debian, Fedora, CentOS, RHEL)

- **Python**: Version 3.8 or higher

- **Git**: Installed and configured

- **NVIDIA GPU**: CUDA-enabled NVIDIA GPU for optimal performance

- **CUDA Toolkit**: Compatible with your GPU and installed

- **FFmpeg**: Installed and added to system `PATH`

- **Internet Connection**: Stable connection for downloading dependencies


### <a name="automated-installation"></a>Automated Installation


For ease of installation, an automated setup script is provided for **Windows** users. This script handles the installation of necessary dependencies and sets up the environment.


#### **Windows Automated Installation**


1. **Download Python Installer**:

  - Visit the [Python Downloads](https://www.python.org/ftp/python/3.10.9/python-3.10.9-amd64.exe) page.

  - Download the **Python 3.10.9** installer for Windows.


2. **Install Python**:

  - **Run the Installer**:

    - Double-click the downloaded `python-3.10.9-amd64.exe` file.

    - **Important**: **Check the box "Add Python to PATH"** at the bottom of the installer window.

    - Click **"Install Now"**.

  - **Verify Installation**:

    - Open **Command Prompt**.

    - Run:

      ```bash

      python --version

      ```

      - You should see `Python 3.10.9`.


3. **Download the Repository**:

  - Visit the [TemporalPromptGenerator GitHub Repository](https://github.com/TemporalLabsLLC-SOL/TemporalPromptGenerator).

  - Click on the **"Code"** button and select **"Download ZIP"**.

  - Extract the downloaded ZIP file to your desired location, e.g., `C:\TemporalPromptEngine`.


4. **Pre-install Essential Tools**:

  - **Install Git**:

    - Download Git from [Git for Windows](https://git-scm.com/download/win).

    - Run the installer with default settings.

    - **Verify Installation**:

      - Open **Command Prompt**.

      - Run:

        ```bash

        git --version

        ```

        - You should see the installed Git version.

  - **Install 7-Zip (if needed)**:

    - Download 7-Zip from [7-Zip Download](https://www.7-zip.org/download.html).

    - Run the installer and follow the on-screen instructions.

  - **Install WinRAR (optional)**:

    - Download WinRAR from [WinRAR Download](https://www.win-rar.com/download.html).

    - Run the installer and follow the on-screen instructions.


5. **Pre-install Additional Dependencies**:

  - **Install Ollama**:

    - Download Ollama from [Ollama Setup](https://ollama.com/download/OllamaSetup.exe).

    - Run the installer and follow the on-screen instructions.

  - **Install CUDA Toolkit**:

    - Ensure you have an NVIDIA GPU with the latest drivers.

    - Download the CUDA Toolkit from [CUDA Toolkit Download](https://developer.nvidia.com/cuda-11.8.0-download-archive).

    - Run the installer and follow the on-screen instructions.

  - **Install FFMPEG**:

    - `https://www.geeksforgeeks.org/how-to-install-ffmpeg-on-windows/`


6. **Run the Setup Script**:

  - **Open Command Prompt**:

    - Navigate to the extracted `TemporalPromptEngine` directory:

      ```bash

      cd C:\TemporalPromptEngine-main

      ```

      (OR Whatever you extracted it as)


  - Navigate to the respective `WINDOWS or LINUX` folder depending on your system. ONLY WINDOWS WORKS RELIABLY RIGHT NOW.

    - ```bash

      cd WINDOWS

      ```


      OR


      ```bash

      cd LINUX

      ```

  - **Run the Setup Script**:

    ```bash

    python SETUP.py

    ```

  - **Follow On-Screen Prompts**:

    - The script will automatically set up the environment, install necessary packages, and configure settings.

    - **IMPORTANT**: If the script prompts for a restart, **RESTART YOUR COMPUTER** to apply changes.


**Activate the Virtual Environment**:

  - The `SETUP.py` script creates a virtual environment named `TemporalPromptEngineEnv` within the project directory.

  - **Activate the Virtual Environment**:

    - In **Terminal**, navigate to the project directory if not already there:

      ```bash

      cd ~/TemporalPromptGenerator

      ```

    - Activate the virtual environment:

      ```bash

      source TemporalPromptEngineEnv/bin/activate

      ```

    - **Confirmation**: You should see `(TemporalPromptEngineEnv)` prefixed in your terminal, indicating that the virtual environment is active.


**Launch the Temporal Prompt Engine**:

  - With the virtual environment activated, run the main script:

    ```bash

    python3 TemporalPromptEngine.py

    ```

  - The application will launch, guiding you through the initial setup.



<a name="4-quick-start-guide"></a>


## 4. Quick Start Guide


Follow these simple steps to begin creating your immersive video and audio scenes:


1. **Activate the Virtual Environment**:

  - **Windows**:

    - Open **Command Prompt**.

    - Navigate to the `TemporalPromptEngine` directory:

      ```bash

      cd C:\TemporalPromptEngine

      ```

    - Activate the virtual environment:

      ```bash

      TemporalPromptEngineEnv\Scripts\activate

      ```

    - **Confirmation**: You should see `(TemporalPromptEngineEnv)` prefixed in your command prompt.

  - **Linux**:

    - Open **Terminal**.

    - Navigate to the `TemporalPromptEngine` directory:

      ```bash

      cd ~/TemporalPromptGenerator

      ```

    - Activate the virtual environment:

      ```bash

      source TemporalPromptEngineEnv/bin/activate

      ```

    - **Confirmation**: You should see `(TemporalPromptEngineEnv)` prefixed in your terminal.


2. **Launch the Application**:

  - With the virtual environment activated, run the main script.

  - **Windows**:

    ```bash

    python TemporalPromptEngine.py

    ```

  - **Linux**:

    ```bash

    python3 TemporalPromptEngine.py

    ```

  - The application will launch, guiding you through the initial setup.


3. **Enter Your Scene Concept**:

  - Input your creative idea or scene description (up to 450 characters).

  - **Example**:

    > "A futuristic cityscape bustling with flying cars, towering skyscrapers adorned with holographic advertisements, under the glow of a setting sun."


4. **Configure Video Prompt Options**:

  - Select standard film terms through dropdown menus.

  - **Example Settings**:

    - **Theme**: Sci-Fi

    - **Art Style**: Cyberpunk

    - **Lighting**: Neon Lighting

    - **Framing**: Wide Shot

    - **Camera Movement**: Tracking Shot

    - **Time of Day**: Dusk

    - **Decade**: 1980s

    - **Camera**: Panavision Panaflex

    - **Lens**: Anamorphic

    - **Resolution**: 4K UHD

  - **Note**: Selecting the **1980s** decade filters the available cameras to those from that era, and choosing the **Anamorphic** lens influences the aspect ratio and cinematic feel of your scene.


5. **Generate Video Prompts**:

  - Click **Generate Video Prompts**.

  - The engine transforms your inputs into detailed video prompts ready for generation.


6. **Configure Audio Prompt Options**:

  - Tailor your soundscape using dropdowns and input fields.

  - **Example Settings**:

    - **Exclude Music**: Enabled (focus on ambient sounds)

    - **Holiday Mode**: Disabled

    - **Specific Modes**: SoundScape Mode

    - **Layer Intensity**: High


7. **Generate Audio Prompts**:

  - Click **Generate Audio Prompts** to create adaptive audio that complements your visuals.


8. **Generate Sound Effects**:

  - Specify the duration of each sound effect.

  - Click **Generate Sound Effects** to produce layered audio, enriching your scenes.


9. **Harnessing the Power of ComfyUI**:

  - Use the included [ComfyUI Workflow](https://github.com/TemporalLabsLLC-SOL/TemporalPromptGenerator) to generate your videos.

  - The engine seamlessly integrates with ComfyUI, allowing you to take the generated video prompts and produce stunning visuals with ease.


10. **Combine Video and Audio**:

   - Use the **COMBINE** feature to merge your video and audio into a seamless, immersive experience.


11. **Save and Export**:

   - All media and prompts are saved in your designated output directory for easy access and future use.


---


<a name="5-api-key-setup"></a>


## 5. API Key Setup


Unlock the full potential of the Temporal Prompt Engine by setting up your API key.


### **HuggingFace API Key**


- **Purpose**: Enables advanced AI-driven prompt generation for both video and audio.

- **Setup**:

 1. **Sign Up at HuggingFace**:

    - Visit [HuggingFace Sign Up](https://huggingface.co/join) and create an account if you don't have one.

 2. **Generate API Key**:

    - After logging in, navigate to your account settings by clicking on your profile picture and selecting **"Settings"**.

    - Click on **"Access Tokens"** in the sidebar.

    - Click **"New Token"**, name it (e.g., "TemporalPromptEngine"), and set the scope to **"Read"**.

    - Click **"Generate"** and **copy your API key**.

 3. **Enter API Key in the Application**:

    - Run the Temporal Prompt Engine.

    - When prompted, paste your HuggingFace API key.

    - The application will store it securely for future use.


**Note**: Your API keys are stored securely, and you only need to enter them once.


---


<a name="6-harnessing-the-power-of-comfyui"></a>


## 6. Harnessing the Power of ComfyUI


The Temporal Prompt Engine is designed to work seamlessly with [ComfyUI](https://comfyui.com/), a powerful tool for AI-based video generation.


- **Integration**: After generating your video prompts, use the included [ComfyUI Workflow](https://github.com/TemporalLabsLLC-SOL/TemporalPromptGenerator) to produce your videos.

- **Process**:

 - **Import Prompts**: The engine exports prompts in a format compatible with ComfyUI.

 - **Generate Videos**: Utilize ComfyUI's advanced capabilities to bring your scenes to life.

 - **Refinement**: Iterate on your prompts and settings based on the outputs, refining your vision.


By combining the strengths of the Temporal Prompt Engine and ComfyUI, you unlock a workflow that is both powerful and user-friendly, streamlining the path from concept to creation.


---


<a name="7-join-the-temporal-labs-journey"></a>


## 7. Join the Temporal Labs Journey


I am a lone futurist on a mission, dedicated to pushing the boundaries of what's possible in AI and technology. Temporal Labs LLC is not just about tools—it's about fostering a community of innovators, developers, artists, and visionaries.


### **Seeking Collaborators and Support**


- **Investors**: Partner with me to accelerate innovation and bring groundbreaking ideas to life.

- **Developers and Students**: Join the journey, contribute to cutting-edge projects, and grow alongside like-minded individuals.

- **Clients and Associates**: Let's work together to create custom solutions that meet your unique needs.

- **Supporters**: Your encouragement and support fuel the mission.


### **Beyond the Temporal Prompt Engine**


Temporal Labs LLC offers a range of AI and tech services across software and hardware. Whether it's developing custom AI models, consulting on technological strategies, or crafting innovative hardware solutions, I'm committed to delivering excellence.


---


<a name="8-donations-and-support"></a>


## 8. Donations and Support


Your support helps continue the mission of innovation and the development of tools like the Temporal Prompt Engine. If you find this tool valuable, consider making a donation.


### **Crypto Donations**


- **Ethereum (ETH)**:

 - Address: `0x5616b3415ED6Ea7005595eF144A2054d4cD5767B`


- **Bitcoin (BTC)**:

 - Address: `bc1qpsfn8a7cs75fxxwv3ax7gtnurm44n5x2fmh59c`


- **Solana (SOL)**:

 - Address: `FVPGxfGT7QWfQEWvXpFkwdgiiKFM3VdvzNG6mEmX8pgi`


- **Litecoin (LTC)**:

 - Address: `ltc1qwlyjz8aymy9uagqhht5a4kaq06kmv58dxlzyww`


- **Dogecoin (DOGE)**:

 - Address: `DAeWAroHCy8nXCoUsobderPRSNXNu1WY34`


*You can copy any of these addresses to your clipboard for easy transfer.*


### **Venmo**


- **Venmo ID**: `@Utah-DM`


---


<a name="9-contact"></a>


## 9. Contact


For questions, support, or to discuss collaboration opportunities, I'd love to hear from you:


- **Email**: [[email protected]](mailto:[email protected])

- **Phone**: +1-385-222-9920


---


<a name="10-acknowledgments"></a>


## 10. Acknowledgments


The development of the Temporal Prompt Engine leverages a variety of open-source tools and libraries:


- **[Git](https://git-scm.com/)**: Version control system.

- **[Python](https://www.python.org/)**: Programming language.

- **[FFmpeg](https://ffmpeg.org/)**: Multimedia framework.

- **[Ollama](https://ollama.com/)**: AI-powered tools.

- **[AudioLDM2](https://github.com/haoheliu/AudioLDM2)**: Audio generation model.

- **[ComfyUI](https://comfyui.com/)**: User interface for AI-based video generation.

- **[HuggingFace](https://huggingface.co/)**: AI models and APIs.


A special thanks to the developers and communities behind these tools for making innovative projects like the Temporal Prompt Engine possible.


---


**Embark on the Temporal Labs Journey**


This is more than just a tool—it's a gateway to infinite creative possibilities. The Temporal Prompt Engine is crafted to inspire and empower, offering you a canvas limited only by your imagination. With its extensive features and intuitive design, creating visually stunning and sonically rich scenes has never been more accessible.


Join me in exploring the future of AI-driven media creation. Your support fuels innovation, and I'm excited to see what you'll create.


---


Visit us at [www.TemporalLab.com](http://www.TemporalLab.com)


---


*Temporal Labs LLC is committed to ethical AI development, focusing on open-source solutions and community-driven innovation. By choosing the Temporal Prompt Engine, you're not just using a tool—you're joining a movement towards responsible and groundbreaking technological advancement.*

---

Feel free to alter the example loaded in the batch file node to whatever length you need.

It determines how many frames to do on the back-end. It defaults for higher quality right now I believe.

COGVIDEOX-5b STEPS QUALITY REFERENCE GUIDE

STEPS BALANCE IS KEY TO NOT WASTING TIME RESOURCE

EXAMPLE SETTINGS

8FPS @ 49 Frames Total

CFG 5.0

Seed - 806286757407563

PROMPT USED

positive:a high-resolution aerial view of Orem in the early 2000s, capturing the city’s growth into a bustling suburban hub. The scene shows the development of new residential neighborhoods, expanded commercial areas, and the growth of Utah Valley University as a major educational institution. The camera continues to lower, providing a closer view of the city’s infrastructure, with modern cars and public transportation visible on the streets. The lighting is bright and clear, reflecting the optimistic start of a new millennium.
negative:text, watermark, any inconsistencies in early 2000s architecture, vehicles, or infrastructure, harsh or artificial lighting, abrupt transitions, low-resolution imagery, any visual glitches or artifacts that detract from the historical accuracy.

5 STEPS

11 STEPS

16 STEPS

22 STEPS

27 STEPS

33 STEPS

67 STEPS

100 STEPS

SYSTEM SPECIFICATIONS

Python Version:

Python 3.10.14

GPU Information:

Mon Sep  2 05:53:16 2024      

+---------------------------------------------------------------------------------------+

| NVIDIA-SMI 535.183.01             Driver Version: 535.183.01   CUDA Version: 12.2     |

|-----------------------------------------+----------------------+----------------------+

| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |

| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |

|                                         |                      |               MIG M. |

|=========================================+======================+======================|

|   0  NVIDIA L40S                    On  | 00000000:34:00.0 Off |                    0 |

| N/A   39C    P0              79W / 350W |  17113MiB / 46068MiB |      0%      Default |

|                                         |                      |                  N/A |

+-----------------------------------------+----------------------+----------------------+

                                                                                       

+---------------------------------------------------------------------------------------+

| Processes:                                                                            |

|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |

|        ID   ID                                                             Usage      |

|=======================================================================================|

|    0   N/A  N/A      2389      C   ...ubuntu/venvs/comfyui_env/bin/python    17100MiB |

+---------------------------------------------------------------------------------------+

IDEAL STEPS PER TIME ALLOTMENT VERSUS PROMPTS PER TXT FILE

USING ABOVE SPECIFICATIONS

MORE INFORMATION LINK COMING SOON

Discussion

(No comments yet)

Loading...

Author

2
966
12
10.1K

No reviews yet

  • - latest (a year ago)

  • - v20240903-212721

  • - v20240903-210247

  • - v20240902-094829

  • - v20240902-081833

Primitive Nodes (8)

CogVideoDecode (1)

CogVideoSampler (1)

CogVideoTextEncode (2)

DownloadAndLoadCogVideoModel (1)

SimpleMathFloat+ (1)

SimpleMathInt+ (2)

Custom Nodes (6)

ComfyUI

  • - CLIPLoader (1)

ComfyUI Essentials

  • - SimpleMath+ (1)

  • - LoadPromptsFromFile //Inspire (1)

  • - UnzipPrompt //Inspire (1)

  • - VHS_VideoCombine (1)

  • - GetImageSizeAndCount (1)

Checkpoints (0)

LoRAs (0)