commit

.gitignore

@@ -0,0 +1,49 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+ENV/
+env/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Dataset cache
+.cache/
+cached/
+
+# Testing
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Hugging Face
+dataset_infos.json.lock

README.md

@@ -0,0 +1,273 @@
+---
+language:
+- en
+license: cc0-1.0
+size_categories:
+- n<1K
+task_categories:
+- other
+pretty_name: Multimodal AI Taxonomy
+tags:
+- multimodal
+- taxonomy
+- ai-models
+- modality-mapping
+- computer-vision
+- audio
+- video-generation
+- image-generation
+---
+
+# Multimodal AI Taxonomy
+
+A comprehensive, structured taxonomy for mapping multimodal AI model capabilities across input and output modalities.
+
+## Dataset Description
+
+This dataset provides a systematic categorization of multimodal AI capabilities, enabling users to:
+- Navigate the complex landscape of multimodal AI models
+- Filter models by specific input/output modality combinations
+- Understand the nuanced differences between similar models (e.g., image-to-video with/without audio, with/without lip sync)
+- Discover models that match specific use case requirements
+
+### Dataset Summary
+
+The taxonomy organizes multimodal AI capabilities by:
+- **Output modality** (video, audio, image, text, 3D models)
+- **Operation type** (creation vs. editing)
+- **Detailed characteristics** (lip sync, audio generation method, motion type, etc.)
+- **Maturity level** (experimental, emerging, mature)
+- **Platform availability** and example models
+
+### Supported Tasks
+
+This is a reference taxonomy dataset for:
+- Model discovery and filtering
+- Understanding multimodal AI capabilities
+- Research into multimodal AI landscape
+- Building model selection tools
+
+## Dataset Structure
+
+The taxonomy is organized as a hierarchical folder structure:
+
+```
+taxonomy/
+├── schema.json                          # Common schema definition
+├── README.md                            # Taxonomy documentation
+├── video-generation/
+│   ├── creation/modalities.json
+│   └── editing/modalities.json
+├── audio-generation/
+│   ├── creation/modalities.json
+│   └── editing/modalities.json
+├── image-generation/
+│   ├── creation/modalities.json
+│   └── editing/modalities.json
+├── text-generation/
+│   ├── creation/modalities.json
+│   └── editing/modalities.json
+└── 3d-generation/
+    ├── creation/modalities.json
+    └── editing/modalities.json
+```
+
+### Data Instances
+
+Each modality entry contains:
+
+```json
+{
+  "id": "img-to-vid-lipsync-text",
+  "name": "Image to Video (Lip Sync from Text)",
+  "input": {
+    "primary": "image",
+    "secondary": ["text"]
+  },
+  "output": {
+    "primary": "video",
+    "audio": true,
+    "audioType": "speech"
+  },
+  "characteristics": {
+    "processType": "synthesis",
+    "audioGeneration": "text-to-speech",
+    "audioPrompting": "text-based",
+    "lipSync": true,
+    "lipSyncMethod": "generated-from-text",
+    "motionType": "facial"
+  },
+  "metadata": {
+    "maturityLevel": "mature",
+    "commonUseCases": [
+      "Avatar creation",
+      "Character animation from portrait",
+      "Marketing personalization"
+    ],
+    "platforms": ["Replicate", "FAL AI", "HeyGen"],
+    "exampleModels": ["Wav2Lip", "SadTalker", "DreamTalk"]
+  }
+}
+```
+
+### Data Fields
+
+**Top-level file fields:**
+- `fileType`: Always "multimodal-ai-taxonomy"
+- `outputModality`: The primary output type (video, audio, image, text, 3d-model)
+- `operationType`: Either "creation" or "editing"
+- `description`: Human-readable description of the file contents
+- `modalities`: Array of modality objects
+
+**Modality object fields:**
+- `id` (string): Unique identifier in kebab-case
+- `name` (string): Human-readable name
+- `input` (object):
+  - `primary` (string): Main input modality
+  - `secondary` (array): Additional optional inputs
+- `output` (object):
+  - `primary` (string): Main output modality
+  - `audio` (boolean): Whether audio is included (for video outputs)
+  - `audioType` (string): Type of audio (speech, music, ambient, etc.)
+- `characteristics` (object): Modality-specific features (varies by type)
+- `metadata` (object):
+  - `maturityLevel` (string): experimental, emerging, or mature
+  - `commonUseCases` (array): Typical use cases
+  - `platforms` (array): Platforms supporting this modality
+  - `exampleModels` (array): Example model implementations
+- `relationships` (object, optional): Links to related modalities
+
+### Data Splits
+
+This dataset is provided as a complete reference taxonomy without splits.
+
+## Dataset Creation
+
+### Curation Rationale
+
+The rapid development of multimodal AI has created a complex landscape with hundreds of model variants. Platforms like Replicate and FAL AI offer numerous models that differ not just in parameters or resolution, but in fundamental modality support. For example, among 20+ image-to-video models, some generate silent video, others add ambient audio, and some include lip-synced speech - but these differences aren't easily filterable.
+
+This taxonomy addresses the need for:
+1. **Systematic categorization** of multimodal capabilities
+2. **Fine-grained filtering** beyond basic input/output types
+3. **Discovery** of models matching specific use cases
+4. **Understanding** of the multimodal AI landscape
+
+### Source Data
+
+The taxonomy is curated from:
+- Public AI model platforms (Replicate, FAL AI, HuggingFace, RunwayML, etc.)
+- Research papers and model documentation
+- Community knowledge and testing
+- Direct platform API exploration
+
+### Annotations
+
+All entries are manually curated and categorized based on model documentation, testing, and platform specifications.
+
+## Considerations for Using the Data
+
+### Social Impact
+
+This dataset is designed to:
+- Democratize access to understanding multimodal AI capabilities
+- Enable better model selection for specific use cases
+- Support research into multimodal AI trends and capabilities
+
+### Discussion of Biases
+
+The taxonomy reflects:
+- Current state of publicly accessible multimodal AI (as of 2025)
+- Platform availability bias toward commercial services
+- Maturity level assessments based on community adoption and stability
+
+### Other Known Limitations
+
+- The field is rapidly evolving; new modalities emerge regularly
+- Platform and model availability changes over time
+- Some experimental modalities may have limited real-world implementations
+- Coverage may be incomplete for niche or newly emerging modalities
+
+## Additional Information
+
+### Dataset Curators
+
+Created and maintained as an open-source project for the multimodal AI community.
+
+### Licensing Information
+
+Creative Commons Zero v1.0 Universal (CC0 1.0) - Public Domain Dedication
+
+### Citation Information
+
+If you use this taxonomy in your research or projects, please cite:
+
+```bibtex
+@dataset{multimodal_ai_taxonomy,
+  title={Multimodal AI Taxonomy},
+  author={Community Contributors},
+  year={2025},
+  publisher={Hugging Face},
+  howpublished={\url{https://huggingface.co/datasets/YOUR_USERNAME/multimodal-ai-taxonomy}}
+}
+```
+
+### Contributions
+
+This is an open-source taxonomy that welcomes community contributions. To add new modalities or update existing entries:
+
+1. Follow the schema defined in `taxonomy/schema.json`
+2. Add entries to the appropriate modality file based on output type and operation
+3. Submit a pull request with clear documentation
+
+For detailed contribution guidelines, see `taxonomy/README.md`.
+
+## Usage Examples
+
+### Loading the Dataset
+
+```python
+from datasets import load_dataset
+
+# Load the entire taxonomy
+dataset = load_dataset("YOUR_USERNAME/multimodal-ai-taxonomy")
+
+# Access specific modality files
+video_creation = dataset["video_generation_creation"]
+audio_editing = dataset["audio_generation_editing"]
+```
+
+### Filtering by Characteristics
+
+```python
+import json
+
+# Find all video generation modalities with lip sync
+with open("taxonomy/video-generation/creation/modalities.json") as f:
+    data = json.load(f)
+
+lipsync_modalities = [
+    m for m in data["modalities"]
+    if m.get("characteristics", {}).get("lipSync") == True
+]
+
+for modality in lipsync_modalities:
+    print(f"{modality['name']}: {modality['id']}")
+```
+
+### Finding Models by Use Case
+
+```python
+# Find mature image generation methods
+with open("taxonomy/image-generation/creation/modalities.json") as f:
+    data = json.load(f)
+
+mature_methods = [
+    m for m in data["modalities"]
+    if m["metadata"]["maturityLevel"] == "mature"
+]
+```
+
+## Contact
+
+For questions, suggestions, or contributions, please open an issue in the dataset repository.

dataset_infos.json

@@ -0,0 +1,96 @@
+{
+  "all": {
+    "description": "Complete taxonomy with all modalities across all output types and operations.",
+    "citation": "@dataset{multimodal_ai_taxonomy,\n  title={Multimodal AI Taxonomy},\n  author={Community Contributors},\n  year={2025},\n  publisher={Hugging Face},\n}",
+    "homepage": "https://huggingface.co/datasets/YOUR_USERNAME/multimodal-ai-taxonomy",
+    "license": "cc0-1.0",
+    "features": {
+      "id": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "name": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "input_primary": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "input_secondary": {
+        "feature": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "output_primary": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "output_audio": {
+        "dtype": "bool",
+        "_type": "Value"
+      },
+      "output_audio_type": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "characteristics": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "metadata_maturity_level": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "metadata_common_use_cases": {
+        "feature": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "metadata_platforms": {
+        "feature": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "metadata_example_models": {
+        "feature": {
+          "dtype": "string",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "relationships": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "file_output_modality": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "file_operation_type": {
+        "dtype": "string",
+        "_type": "Value"
+      }
+    },
+    "supervised_keys": null,
+    "builder_name": "multimodal_ai_taxonomy",
+    "config_name": "all",
+    "version": "1.0.0",
+    "splits": {
+      "train": {
+        "name": "train",
+        "num_bytes": 0,
+        "num_examples": 0,
+        "dataset_name": "multimodal_ai_taxonomy"
+      }
+    },
+    "download_size": 0,
+    "dataset_size": 0
+  }
+}

multimodal-ai-taxonomy.py

@@ -0,0 +1,217 @@
+"""Multimodal AI Taxonomy dataset loading script."""
+
+import json
+import os
+from pathlib import Path
+from typing import Dict, List
+
+import datasets
+
+
+_CITATION = """\
+@dataset{multimodal_ai_taxonomy,
+  title={Multimodal AI Taxonomy},
+  author={Community Contributors},
+  year={2025},
+  publisher={Hugging Face},
+}
+"""
+
+_DESCRIPTION = """\
+A comprehensive, structured taxonomy for mapping multimodal AI model capabilities across input and output modalities.
+This dataset provides a systematic categorization of multimodal AI capabilities, enabling users to navigate the complex
+landscape of multimodal AI models, filter by specific input/output modality combinations, and discover models that match
+specific use case requirements.
+"""
+
+_HOMEPAGE = "https://huggingface.co/datasets/YOUR_USERNAME/multimodal-ai-taxonomy"
+
+_LICENSE = "cc0-1.0"
+
+_URLS = {
+    "schema": "taxonomy/schema.json",
+    "video_generation_creation": "taxonomy/video-generation/creation/modalities.json",
+    "video_generation_editing": "taxonomy/video-generation/editing/modalities.json",
+    "audio_generation_creation": "taxonomy/audio-generation/creation/modalities.json",
+    "audio_generation_editing": "taxonomy/audio-generation/editing/modalities.json",
+    "image_generation_creation": "taxonomy/image-generation/creation/modalities.json",
+    "image_generation_editing": "taxonomy/image-generation/editing/modalities.json",
+    "text_generation_creation": "taxonomy/text-generation/creation/modalities.json",
+    "text_generation_editing": "taxonomy/text-generation/editing/modalities.json",
+    "3d_generation_creation": "taxonomy/3d-generation/creation/modalities.json",
+    "3d_generation_editing": "taxonomy/3d-generation/editing/modalities.json",
+}
+
+
+class MultimodalAITaxonomy(datasets.GeneratorBasedBuilder):
+    """Multimodal AI Taxonomy dataset."""
+
+    VERSION = datasets.Version("1.0.0")
+
+    BUILDER_CONFIGS = [
+        datasets.BuilderConfig(
+            name="all",
+            version=VERSION,
+            description="Complete taxonomy with all modalities",
+        ),
+        datasets.BuilderConfig(
+            name="video_generation_creation",
+            version=VERSION,
+            description="Video generation (creation) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="video_generation_editing",
+            version=VERSION,
+            description="Video generation (editing) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="audio_generation_creation",
+            version=VERSION,
+            description="Audio generation (creation) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="audio_generation_editing",
+            version=VERSION,
+            description="Audio generation (editing) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="image_generation_creation",
+            version=VERSION,
+            description="Image generation (creation) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="image_generation_editing",
+            version=VERSION,
+            description="Image generation (editing) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="text_generation_creation",
+            version=VERSION,
+            description="Text generation (creation) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="text_generation_editing",
+            version=VERSION,
+            description="Text generation (editing) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="3d_generation_creation",
+            version=VERSION,
+            description="3D generation (creation) modalities",
+        ),
+        datasets.BuilderConfig(
+            name="3d_generation_editing",
+            version=VERSION,
+            description="3D generation (editing) modalities",
+        ),
+    ]
+
+    DEFAULT_CONFIG_NAME = "all"
+
+    def _info(self):
+        features = datasets.Features(
+            {
+                "id": datasets.Value("string"),
+                "name": datasets.Value("string"),
+                "input_primary": datasets.Value("string"),
+                "input_secondary": datasets.Sequence(datasets.Value("string")),
+                "output_primary": datasets.Value("string"),
+                "output_audio": datasets.Value("bool"),
+                "output_audio_type": datasets.Value("string"),
+                "characteristics": datasets.Value("string"),  # JSON string for flexibility
+                "metadata_maturity_level": datasets.Value("string"),
+                "metadata_common_use_cases": datasets.Sequence(datasets.Value("string")),
+                "metadata_platforms": datasets.Sequence(datasets.Value("string")),
+                "metadata_example_models": datasets.Sequence(datasets.Value("string")),
+                "relationships": datasets.Value("string"),  # JSON string for flexibility
+                "file_output_modality": datasets.Value("string"),
+                "file_operation_type": datasets.Value("string"),
+            }
+        )
+
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features,
+            homepage=_HOMEPAGE,
+            license=_LICENSE,
+            citation=_CITATION,
+        )
+
+    def _split_generators(self, dl_manager):
+        """Returns SplitGenerators."""
+
+        # Download/locate all files
+        if self.config.name == "all":
+            # Load all modality files
+            config_names = [k for k in _URLS.keys() if k != "schema"]
+        else:
+            # Load only the specified config
+            config_names = [self.config.name]
+
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={
+                    "config_names": config_names,
+                    "dl_manager": dl_manager,
+                },
+            ),
+        ]
+
+    def _generate_examples(self, config_names, dl_manager):
+        """Yields examples from the taxonomy."""
+
+        idx = 0
+        for config_name in config_names:
+            filepath = _URLS[config_name]
+
+            # Read the JSON file
+            with open(dl_manager.download(filepath), encoding="utf-8") as f:
+                data = json.load(f)
+
+            output_modality = data.get("outputModality", "")
+            operation_type = data.get("operationType", "")
+
+            # Process each modality in the file
+            for modality in data.get("modalities", []):
+                # Extract input information
+                input_data = modality.get("input", {})
+                input_primary = input_data.get("primary", "")
+                input_secondary = input_data.get("secondary", [])
+
+                # Extract output information
+                output_data = modality.get("output", {})
+                output_primary = output_data.get("primary", "")
+                output_audio = output_data.get("audio", False)
+                output_audio_type = output_data.get("audioType", "")
+
+                # Extract metadata
+                metadata = modality.get("metadata", {})
+                maturity_level = metadata.get("maturityLevel", "")
+                common_use_cases = metadata.get("commonUseCases", [])
+                platforms = metadata.get("platforms", [])
+                example_models = metadata.get("exampleModels", [])
+
+                # Keep characteristics and relationships as JSON strings for flexibility
+                characteristics = json.dumps(modality.get("characteristics", {}))
+                relationships = json.dumps(modality.get("relationships", {}))
+
+                yield idx, {
+                    "id": modality.get("id", ""),
+                    "name": modality.get("name", ""),
+                    "input_primary": input_primary,
+                    "input_secondary": input_secondary,
+                    "output_primary": output_primary,
+                    "output_audio": output_audio,
+                    "output_audio_type": output_audio_type,
+                    "characteristics": characteristics,
+                    "metadata_maturity_level": maturity_level,
+                    "metadata_common_use_cases": common_use_cases,
+                    "metadata_platforms": platforms,
+                    "metadata_example_models": example_models,
+                    "relationships": relationships,
+                    "file_output_modality": output_modality,
+                    "file_operation_type": operation_type,
+                }
+
+                idx += 1

taxonomy/3d-generation/creation/modalities.json

@@ -0,0 +1,60 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "3d-model",
+  "operationType": "creation",
+  "description": "Modalities for creating 3D model content from text, images, or other inputs",
+  "modalities": [
+    {
+      "id": "text-to-3d",
+      "name": "Text to 3D Model",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "3d-model",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "generationType": "3d-synthesis"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "3D asset generation",
+          "Rapid prototyping",
+          "Game asset creation"
+        ],
+        "platforms": ["Replicate", "Meshy", "3DFY"],
+        "exampleModels": ["Point-E", "Shap-E", "DreamFusion"]
+      }
+    },
+    {
+      "id": "img-to-3d",
+      "name": "Image to 3D Model",
+      "input": {
+        "primary": "image",
+        "secondary": []
+      },
+      "output": {
+        "primary": "3d-model",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "generationType": "3d-reconstruction"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "3D reconstruction",
+          "Object digitization",
+          "Asset creation from photos"
+        ],
+        "platforms": ["Replicate", "Meshy", "Luma AI"],
+        "exampleModels": ["Zero-1-to-3", "Wonder3D"]
+      }
+    }
+  ]
+}

taxonomy/3d-generation/editing/modalities.json

@@ -0,0 +1,7 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "3d-model",
+  "operationType": "editing",
+  "description": "Modalities for editing and transforming existing 3D model content (placeholder for future expansion)",
+  "modalities": []
+}

taxonomy/README.md

@@ -0,0 +1,257 @@
+# Multimodal AI Taxonomy
+
+A comprehensive, folder-based taxonomy for mapping multimodal AI model capabilities across input and output modalities.
+
+## Structure
+
+The taxonomy is organized by **output modality** with subfolders for **creation** vs **editing** operations:
+
+```
+taxonomy/
+├── schema.json                          # Common schema definition for all modality files
+├── video-generation/
+│   ├── creation/
+│   │   └── modalities.json             # Creating video from scratch
+│   └── editing/
+│       └── modalities.json             # Transforming existing video
+├── audio-generation/
+│   ├── creation/
+│   │   └── modalities.json             # Creating audio from scratch
+│   └── editing/
+│       └── modalities.json             # Transforming existing audio
+├── image-generation/
+│   ├── creation/
+│   │   └── modalities.json             # Creating images from scratch
+│   └── editing/
+│       └── modalities.json             # Transforming existing images
+├── text-generation/
+│   ├── creation/
+│   │   └── modalities.json             # Creating text from scratch (future)
+│   └── editing/
+│       └── modalities.json             # Transforming existing text (future)
+└── 3d-generation/
+    ├── creation/
+    │   └── modalities.json             # Creating 3D models
+    └── editing/
+        └── modalities.json             # Transforming 3D models (future)
+```
+
+## Organizational Principles
+
+### 1. Output Modality Organization
+- Folders are organized by **what is being generated/produced**
+- Example: "Image to Video" lives in `video-generation/` because video is the output
+
+### 2. Creation vs Editing
+- **Creation**: Generating new content from scratch (text-to-video, image-to-video, etc.)
+- **Editing**: Modifying existing content (video-to-video, audio-to-audio, image-to-image)
+
+### 3. Multimodal Inputs
+- When multiple inputs are used (e.g., image + audio → video), file placement is determined by the **output** modality
+- Example: "Image + Audio to Video" goes in `video-generation/creation/`
+
+## Schema Definition
+
+All modality JSON files follow a common schema defined in `schema.json`. Each file contains:
+
+```json
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "video|audio|image|text|3d-model",
+  "operationType": "creation|editing",
+  "description": "Human-readable description",
+  "modalities": [
+    // Array of modality objects
+  ]
+}
+```
+
+### Modality Object Structure
+
+Each modality in the array follows this pattern:
+
+```json
+{
+  "id": "unique-kebab-case-id",
+  "name": "Human Readable Name",
+  "input": {
+    "primary": "main-input-type",
+    "secondary": ["additional", "input", "types"]
+  },
+  "output": {
+    "primary": "output-type",
+    "audio": true|false,
+    "audioType": "speech|music|ambient|etc"
+  },
+  "characteristics": {
+    // Flexible object with modality-specific characteristics
+    // See schema.json for available fields
+  },
+  "metadata": {
+    "maturityLevel": "experimental|emerging|mature",
+    "commonUseCases": ["use case 1", "use case 2"],
+    "platforms": ["Platform 1", "Platform 2"],
+    "exampleModels": ["Model 1", "Model 2"]
+  },
+  "relationships": {
+    // Optional: parent/child/related modality IDs
+  }
+}
+```
+
+## For AI Agents
+
+When working with this taxonomy:
+
+### Adding New Modalities
+
+1. Determine the **output modality** (video, audio, image, text, 3d-model)
+2. Determine if it's **creation** (generating new) or **editing** (transforming existing)
+3. Navigate to the appropriate folder (e.g., `video-generation/creation/`)
+4. Add the new modality object to the `modalities` array in `modalities.json`
+5. Follow the schema defined in `schema.json`
+
+### Example: Adding "Text to Audio with Emotion Control"
+
+This generates audio (output) from scratch (creation):
+
+```bash
+# Edit: audio-generation/creation/modalities.json
+# Add to the modalities array:
+{
+  "id": "text-to-audio-emotion-control",
+  "name": "Text to Audio (Emotion Control)",
+  "input": {
+    "primary": "text",
+    "secondary": []
+  },
+  "output": {
+    "primary": "audio",
+    "audioType": "speech"
+  },
+  "characteristics": {
+    "processType": "synthesis",
+    "audioType": "speech",
+    "voiceCloning": false,
+    "emotionControl": true
+  },
+  "metadata": {
+    "maturityLevel": "emerging",
+    "commonUseCases": [
+      "Emotional narration",
+      "Character voice acting",
+      "Interactive storytelling"
+    ],
+    "platforms": ["ElevenLabs", "Experimental"],
+    "exampleModels": []
+  }
+}
+```
+
+### Creating New Modality Categories
+
+If you need to add a new output modality type (e.g., `haptic-generation`):
+
+1. Create the folder structure:
+   ```bash
+   mkdir -p taxonomy/haptic-generation/{creation,editing}
+   ```
+
+2. Create `modalities.json` files in both subfolders:
+   ```json
+   {
+     "fileType": "multimodal-ai-taxonomy",
+     "outputModality": "haptic",
+     "operationType": "creation",
+     "description": "Modalities for creating haptic feedback from various inputs",
+     "modalities": []
+   }
+   ```
+
+3. Update `schema.json` to include "haptic" in the output modality enum
+
+### Querying the Taxonomy
+
+To find modalities matching specific criteria:
+
+1. **By output type**: Navigate to the appropriate folder (e.g., `video-generation/`)
+2. **By operation**: Look in `creation/` or `editing/` subfolder
+3. **By characteristics**: Parse JSON and filter by characteristics fields
+   - Lip sync: `characteristics.lipSync === true`
+   - Audio generation: `output.audio === true`
+   - Maturity: `metadata.maturityLevel === "mature"`
+
+### Filtering Examples
+
+**Use Case 1**: Find all ways to generate video with lip sync from text
+```
+Folder: video-generation/creation/
+Filter: characteristics.lipSync === true
+        AND characteristics.audioPrompting === "text-based"
+Results: img-to-vid-lipsync-text, img-to-vid-lipsync-lora
+```
+
+**Use Case 2**: Find mature image generation methods
+```
+Folder: image-generation/creation/
+Filter: metadata.maturityLevel === "mature"
+Results: text-to-img
+```
+
+**Use Case 3**: Find all audio editing capabilities
+```
+Folder: audio-generation/editing/
+Filter: All modalities in this folder
+Results: audio-to-audio-inpainting, music-to-music-inpainting
+```
+
+## Key Characteristics Fields
+
+Different modality types use different characteristics fields. Common ones include:
+
+- **processType**: synthesis, transformation, inpainting, editing, enhancement, rendering
+- **audioGeneration**: none, synthesized, text-to-speech, reference-based
+- **lipSync**: true/false
+- **motionType**: general, facial, audio-driven, audio-reactive, guided, camera-path
+- **transformationTypes**: array of transformation types for editing operations
+- **maturityLevel**: experimental, emerging, mature
+
+See `schema.json` for the complete list of available fields and their allowed values.
+
+## Extending the Schema
+
+As new modality types emerge:
+
+1. Add new enum values to appropriate fields in `schema.json`
+2. Add new characteristics properties as needed
+3. Document the new fields in the schema's property descriptions
+4. Update this README with examples
+
+## Validation
+
+To validate that a modality file follows the schema:
+
+```bash
+# Using a JSON schema validator
+npm install -g ajv-cli
+ajv validate -s taxonomy/schema.json -d taxonomy/video-generation/creation/modalities.json
+```
+
+## Migration from Original Structure
+
+The original `taxonomy.json` file has been migrated to this folder-based structure:
+
+- Single flat array → Organized by output modality and operation type
+- Categories metadata → Implicit in folder structure
+- All modality objects → Distributed across appropriate modality files
+- Schema enforcement → Explicit schema.json for validation
+
+## Benefits of This Structure
+
+1. **Scalability**: Easy to add new modalities without file bloat
+2. **Organization**: Clear categorization by output and operation type
+3. **Discoverability**: Navigate folder structure to find relevant modalities
+4. **Maintainability**: Each file focuses on a specific domain
+5. **Agent-friendly**: Clear patterns for AI agents to follow when adding/querying
+6. **Extensible**: Easy to add new modality types as folders
+7. **Schema-enforced**: Common pattern ensures consistency

taxonomy/audio-generation/creation/modalities.json

@@ -0,0 +1,89 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "audio",
+  "operationType": "creation",
+  "description": "Modalities for creating audio content from text or other inputs",
+  "modalities": [
+    {
+      "id": "text-to-audio",
+      "name": "Text to Audio",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "audio",
+        "audioType": "general"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioType": "general",
+        "audioCategories": ["speech", "sound-effects", "music", "ambient"]
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Sound effect generation",
+          "Voiceover creation",
+          "Audio asset production"
+        ],
+        "platforms": ["Replicate", "ElevenLabs", "AudioCraft"],
+        "exampleModels": ["AudioGen", "MusicGen"]
+      }
+    },
+    {
+      "id": "text-to-speech",
+      "name": "Text to Speech",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "audio",
+        "audioType": "speech"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioType": "speech",
+        "voiceCloning": false
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Narration",
+          "Accessibility",
+          "Voice assistants"
+        ],
+        "platforms": ["ElevenLabs", "Google Cloud", "Azure", "AWS"],
+        "exampleModels": ["ElevenLabs", "Google WaveNet", "Azure Neural TTS"]
+      }
+    },
+    {
+      "id": "text-to-music",
+      "name": "Text to Music",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "audio",
+        "audioType": "music"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioType": "music",
+        "melodic": true
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Background music generation",
+          "Musical composition",
+          "Soundtrack creation"
+        ],
+        "platforms": ["Replicate", "Stability AI"],
+        "exampleModels": ["MusicGen", "Stable Audio"]
+      }
+    }
+  ]
+}

taxonomy/audio-generation/editing/modalities.json

@@ -0,0 +1,66 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "audio",
+  "operationType": "editing",
+  "description": "Modalities for editing and transforming existing audio content",
+  "modalities": [
+    {
+      "id": "audio-to-audio-inpainting",
+      "name": "Audio to Audio (Inpainting)",
+      "input": {
+        "primary": "audio",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "audio",
+        "audioType": "general"
+      },
+      "characteristics": {
+        "processType": "inpainting",
+        "modification": "selective-editing"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Audio editing",
+          "Sound design",
+          "Audio restoration"
+        ],
+        "platforms": ["Experimental"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "music-to-music-inpainting",
+      "name": "Music to Music (Inpainting)",
+      "input": {
+        "primary": "audio",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "audio",
+        "audioType": "music"
+      },
+      "characteristics": {
+        "processType": "inpainting",
+        "modification": "selective-editing",
+        "melodic": true,
+        "audioSubtype": "music"
+      },
+      "metadata": {
+        "maturityLevel": "experimental",
+        "commonUseCases": [
+          "Music editing",
+          "Compositional modifications",
+          "Arrangement changes"
+        ],
+        "platforms": ["Experimental"],
+        "exampleModels": []
+      },
+      "relationships": {
+        "parent": "audio-to-audio-inpainting",
+        "note": "Music inpainting is a specialized subset of audio inpainting"
+      }
+    }
+  ]
+}

taxonomy/image-generation/creation/modalities.json

@@ -0,0 +1,34 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "image",
+  "operationType": "creation",
+  "description": "Modalities for creating image content from text or other inputs",
+  "modalities": [
+    {
+      "id": "text-to-img",
+      "name": "Text to Image",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "image",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "generationType": "synthesis"
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Concept art generation",
+          "Product mockups",
+          "Marketing assets"
+        ],
+        "platforms": ["Replicate", "Stability AI", "Midjourney", "DALL-E"],
+        "exampleModels": ["Stable Diffusion", "DALL-E 3", "Midjourney"]
+      }
+    }
+  ]
+}

taxonomy/image-generation/editing/modalities.json

@@ -0,0 +1,35 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "image",
+  "operationType": "editing",
+  "description": "Modalities for editing and transforming existing image content",
+  "modalities": [
+    {
+      "id": "img-to-img",
+      "name": "Image to Image",
+      "input": {
+        "primary": "image",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "image",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "transformation",
+        "transformationTypes": ["style-transfer", "enhancement", "editing", "inpainting"]
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Image editing",
+          "Style transfer",
+          "Image enhancement",
+          "Object removal/addition"
+        ],
+        "platforms": ["Replicate", "Stability AI", "Midjourney"],
+        "exampleModels": ["Stable Diffusion img2img", "ControlNet"]
+      }
+    }
+  ]
+}

taxonomy/schema.json

@@ -0,0 +1,291 @@
+{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "title": "Multimodal AI Taxonomy Schema",
+  "version": "1.0.0",
+  "description": "Common schema for all modality JSON arrays in the taxonomy. Each modality file should contain an array of modality objects following this schema.",
+
+  "modalitySchema": {
+    "type": "object",
+    "required": ["id", "name", "input", "output", "characteristics", "metadata"],
+    "properties": {
+      "id": {
+        "type": "string",
+        "description": "Unique identifier for this modality (kebab-case)",
+        "pattern": "^[a-z0-9-]+$"
+      },
+      "name": {
+        "type": "string",
+        "description": "Human-readable name for this modality"
+      },
+      "input": {
+        "type": "object",
+        "required": ["primary"],
+        "properties": {
+          "primary": {
+            "type": "string",
+            "description": "Primary input modality",
+            "enum": ["text", "image", "video", "audio", "music", "3d-model", "lora-model"]
+          },
+          "secondary": {
+            "type": "array",
+            "description": "Additional input modalities that can be provided",
+            "items": {
+              "type": "string",
+              "enum": ["text", "image", "video", "audio", "music", "3d-model", "lora-model"]
+            }
+          }
+        }
+      },
+      "output": {
+        "type": "object",
+        "required": ["primary"],
+        "properties": {
+          "primary": {
+            "type": "string",
+            "description": "Primary output modality",
+            "enum": ["text", "image", "video", "audio", "music", "3d-model"]
+          },
+          "audio": {
+            "type": "boolean",
+            "description": "Whether audio is included in the output (for video outputs)"
+          },
+          "audioType": {
+            "type": "string",
+            "description": "Type of audio in the output",
+            "enum": ["speech", "music", "ambient", "synchronized", "original", "general"]
+          }
+        }
+      },
+      "characteristics": {
+        "type": "object",
+        "description": "Specific characteristics and capabilities of this modality. Fields vary based on modality type.",
+        "properties": {
+          "processType": {
+            "type": "string",
+            "description": "Type of processing being performed",
+            "enum": ["synthesis", "transformation", "inpainting", "editing", "enhancement", "rendering"]
+          },
+          "audioGeneration": {
+            "type": "string",
+            "description": "How audio is generated",
+            "enum": ["none", "synthesized", "text-to-speech", "reference-based"]
+          },
+          "audioPrompting": {
+            "type": "string",
+            "description": "How audio generation is prompted",
+            "enum": ["text-based", "audio-reference", "none"]
+          },
+          "lipSync": {
+            "type": "boolean",
+            "description": "Whether lip sync is supported/generated"
+          },
+          "lipSyncMethod": {
+            "type": "string",
+            "description": "Method used for lip sync",
+            "enum": ["generated-from-text", "audio-driven"]
+          },
+          "motionType": {
+            "type": "string",
+            "description": "Type of motion in video output",
+            "enum": ["general", "facial", "audio-driven", "audio-reactive", "guided", "camera-path"]
+          },
+          "audioCharacteristics": {
+            "type": "array",
+            "description": "Specific characteristics of generated audio",
+            "items": {
+              "type": "string",
+              "enum": ["background", "environmental", "atmospheric", "melodic", "rhythmic"]
+            }
+          },
+          "transformationTypes": {
+            "type": "array",
+            "description": "Types of transformations supported",
+            "items": {
+              "type": "string",
+              "enum": ["style-transfer", "enhancement", "editing", "inpainting", "motion-modification", "object-editing"]
+            }
+          },
+          "preserveAudio": {
+            "type": "boolean",
+            "description": "Whether original audio is preserved in video transformations"
+          },
+          "audioHandling": {
+            "type": "string",
+            "description": "How audio is handled during processing",
+            "enum": ["passthrough", "removed", "generated", "modified"]
+          },
+          "characterReference": {
+            "type": "string",
+            "description": "Method for character reference/consistency",
+            "enum": ["lora", "image", "video"]
+          },
+          "audioVideoSync": {
+            "type": "boolean",
+            "description": "Whether audio and video are synchronized/coherent"
+          },
+          "audioVisualization": {
+            "type": "boolean",
+            "description": "Whether visuals are generated from audio"
+          },
+          "audioType": {
+            "type": "string",
+            "description": "Type of audio output",
+            "enum": ["speech", "music", "ambient", "general", "sound-effects"]
+          },
+          "audioCategories": {
+            "type": "array",
+            "description": "Categories of audio that can be generated",
+            "items": {
+              "type": "string",
+              "enum": ["speech", "sound-effects", "music", "ambient"]
+            }
+          },
+          "voiceCloning": {
+            "type": "boolean",
+            "description": "Whether voice cloning is supported"
+          },
+          "melodic": {
+            "type": "boolean",
+            "description": "Whether output is melodic (for music/audio)"
+          },
+          "audioSubtype": {
+            "type": "string",
+            "description": "Specific subtype of audio",
+            "enum": ["music", "speech", "effects", "ambient"]
+          },
+          "modification": {
+            "type": "string",
+            "description": "Type of modification being performed",
+            "enum": ["selective-editing", "enhancement", "restoration", "transformation"]
+          },
+          "generationType": {
+            "type": "string",
+            "description": "Type of generation process",
+            "enum": ["synthesis", "3d-synthesis", "3d-reconstruction"]
+          },
+          "renderType": {
+            "type": "string",
+            "description": "Type of rendering process",
+            "enum": ["3d-rendering", "2d-rendering"]
+          },
+          "guidanceType": {
+            "type": "string",
+            "description": "Type of guidance provided to the model",
+            "enum": ["text-only", "text-and-visual", "audio-driven", "multimodal"]
+          }
+        }
+      },
+      "metadata": {
+        "type": "object",
+        "required": ["maturityLevel", "commonUseCases"],
+        "properties": {
+          "maturityLevel": {
+            "type": "string",
+            "description": "How established/stable this modality is",
+            "enum": ["experimental", "emerging", "mature"]
+          },
+          "commonUseCases": {
+            "type": "array",
+            "description": "Common use cases for this modality",
+            "items": {
+              "type": "string"
+            },
+            "minItems": 1
+          },
+          "platforms": {
+            "type": "array",
+            "description": "Platforms/services that support this modality",
+            "items": {
+              "type": "string"
+            }
+          },
+          "exampleModels": {
+            "type": "array",
+            "description": "Example models that implement this modality",
+            "items": {
+              "type": "string"
+            }
+          }
+        }
+      },
+      "relationships": {
+        "type": "object",
+        "description": "Relationships to other modalities (optional)",
+        "properties": {
+          "parent": {
+            "type": "string",
+            "description": "Parent modality ID if this is a specialization"
+          },
+          "children": {
+            "type": "array",
+            "description": "Child modality IDs that specialize this modality",
+            "items": {
+              "type": "string"
+            }
+          },
+          "related": {
+            "type": "array",
+            "description": "Related modality IDs",
+            "items": {
+              "type": "string"
+            }
+          },
+          "note": {
+            "type": "string",
+            "description": "Additional notes about the relationship"
+          }
+        }
+      }
+    }
+  },
+
+  "fileStructure": {
+    "description": "Each modality JSON file should be an object with metadata and an array of modalities",
+    "type": "object",
+    "required": ["fileType", "outputModality", "operationType", "modalities"],
+    "properties": {
+      "fileType": {
+        "type": "string",
+        "const": "multimodal-ai-taxonomy",
+        "description": "Identifier for this file type"
+      },
+      "outputModality": {
+        "type": "string",
+        "description": "The primary output modality this file describes",
+        "enum": ["video", "audio", "image", "text", "3d-model"]
+      },
+      "operationType": {
+        "type": "string",
+        "description": "Whether this file describes creation or editing operations",
+        "enum": ["creation", "editing"]
+      },
+      "description": {
+        "type": "string",
+        "description": "Human-readable description of what this file contains"
+      },
+      "modalities": {
+        "type": "array",
+        "description": "Array of modality objects following the modalitySchema",
+        "items": {
+          "$ref": "#/modalitySchema"
+        }
+      }
+    }
+  },
+
+  "usageInstructions": {
+    "description": "Instructions for AI agents working with this taxonomy",
+    "guidelines": [
+      "Each JSON file in the taxonomy represents a collection of related modalities",
+      "All modality objects must follow the modalitySchema defined above",
+      "Files are organized by output modality (video-generation, audio-generation, etc.)",
+      "Within each modality folder, files are separated into creation/ and editing/ subfolders",
+      "Creation: Generating content from scratch (text-to-video, image-to-video, etc.)",
+      "Editing: Modifying existing content (video-to-video, audio-to-audio, image-to-image, etc.)",
+      "For multimodal inputs, file placement is determined by OUTPUT modality",
+      "The characteristics object is flexible - use appropriate fields for each modality type",
+      "Maintain consistent IDs using kebab-case format",
+      "Add new enum values to the schema as new modality types emerge"
+    ]
+  }
+}

taxonomy/text-generation/creation/modalities.json

@@ -0,0 +1,7 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "text",
+  "operationType": "creation",
+  "description": "Modalities for creating text content from various inputs (placeholder for future expansion)",
+  "modalities": []
+}

taxonomy/text-generation/editing/modalities.json

@@ -0,0 +1,7 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "text",
+  "operationType": "editing",
+  "description": "Modalities for editing and transforming existing text content (placeholder for future expansion)",
+  "modalities": []
+}

taxonomy/video-generation/creation/modalities.json

@@ -0,0 +1,328 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "video",
+  "operationType": "creation",
+  "description": "Modalities for creating video content from various input types (text, images, audio, 3D models)",
+  "modalities": [
+    {
+      "id": "img-to-vid-no-audio",
+      "name": "Image to Video (No Audio)",
+      "input": {
+        "primary": "image",
+        "secondary": []
+      },
+      "output": {
+        "primary": "video",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "none",
+        "lipSync": false,
+        "motionType": "general"
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Static image animation",
+          "Product visualization",
+          "Concept previsualization"
+        ],
+        "platforms": ["Replicate", "FAL AI", "Stability AI"],
+        "exampleModels": ["Stable Video Diffusion", "AnimateDiff"]
+      }
+    },
+    {
+      "id": "img-to-vid-ambient-audio",
+      "name": "Image to Video (Ambient Audio)",
+      "input": {
+        "primary": "image",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "ambient"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "synthesized",
+        "audioPrompting": "text-based",
+        "lipSync": false,
+        "motionType": "general",
+        "audioCharacteristics": ["background", "environmental", "atmospheric"]
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Scene ambiance creation",
+          "Marketplace atmosphere",
+          "Environmental storytelling"
+        ],
+        "platforms": ["FAL AI", "Experimental"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "img-to-vid-lipsync-text",
+      "name": "Image to Video (Lip Sync from Text)",
+      "input": {
+        "primary": "image",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "speech"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "text-to-speech",
+        "audioPrompting": "text-based",
+        "lipSync": true,
+        "lipSyncMethod": "generated-from-text",
+        "motionType": "facial"
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Avatar creation",
+          "Character animation from portrait",
+          "Marketing personalization"
+        ],
+        "platforms": ["Replicate", "FAL AI", "HeyGen"],
+        "exampleModels": ["Wav2Lip", "SadTalker", "DreamTalk"]
+      }
+    },
+    {
+      "id": "img-to-vid-lipsync-audio",
+      "name": "Image to Video (Lip Sync from Audio)",
+      "input": {
+        "primary": "image",
+        "secondary": ["audio"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "speech"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "reference-based",
+        "audioPrompting": "audio-reference",
+        "lipSync": true,
+        "lipSyncMethod": "audio-driven",
+        "motionType": "facial"
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "Voice cloning with video",
+          "Dubbing and localization",
+          "Podcast video generation"
+        ],
+        "platforms": ["Replicate", "FAL AI"],
+        "exampleModels": ["Wav2Lip", "SadTalker"]
+      }
+    },
+    {
+      "id": "img-to-vid-lipsync-lora",
+      "name": "Image to Video (Lip Sync with LoRA Character)",
+      "input": {
+        "primary": "image",
+        "secondary": ["text", "lora-model"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "speech"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "text-to-speech",
+        "audioPrompting": "text-based",
+        "lipSync": true,
+        "lipSyncMethod": "generated-from-text",
+        "characterReference": "lora",
+        "motionType": "facial"
+      },
+      "metadata": {
+        "maturityLevel": "experimental",
+        "commonUseCases": [
+          "Consistent character animation",
+          "Brand mascot videos",
+          "Personalized avatars"
+        ],
+        "platforms": ["Specialized services"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "text-to-vid-no-audio",
+      "name": "Text to Video (No Audio)",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "video",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "none",
+        "motionType": "general"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Concept visualization",
+          "Storyboarding",
+          "Creative exploration"
+        ],
+        "platforms": ["Replicate", "FAL AI", "RunwayML"],
+        "exampleModels": ["ModelScope", "ZeroScope"]
+      }
+    },
+    {
+      "id": "text-to-vid-with-audio",
+      "name": "Text to Video (With Audio)",
+      "input": {
+        "primary": "text",
+        "secondary": []
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "synchronized"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "synthesized",
+        "audioPrompting": "text-based",
+        "audioVideoSync": true,
+        "motionType": "general"
+      },
+      "metadata": {
+        "maturityLevel": "experimental",
+        "commonUseCases": [
+          "Complete scene generation",
+          "Multimedia storytelling"
+        ],
+        "platforms": ["Experimental"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "audio-to-vid",
+      "name": "Audio to Video",
+      "input": {
+        "primary": "audio",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "original"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioVisualization": true,
+        "motionType": "audio-reactive"
+      },
+      "metadata": {
+        "maturityLevel": "experimental",
+        "commonUseCases": [
+          "Music visualization",
+          "Audio-reactive art",
+          "Podcast video generation"
+        ],
+        "platforms": ["Experimental"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "multimodal-img-audio-to-vid",
+      "name": "Image + Audio to Video",
+      "input": {
+        "primary": "image",
+        "secondary": ["audio"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "original"
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "audioGeneration": "reference-based",
+        "motionType": "audio-driven",
+        "lipSync": false
+      },
+      "metadata": {
+        "maturityLevel": "experimental",
+        "commonUseCases": [
+          "Audio-driven animation",
+          "Dance video generation",
+          "Music-driven motion"
+        ],
+        "platforms": ["Experimental"],
+        "exampleModels": []
+      }
+    },
+    {
+      "id": "multimodal-text-img-to-vid",
+      "name": "Text + Image to Video",
+      "input": {
+        "primary": "text",
+        "secondary": ["image"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "synthesis",
+        "guidanceType": "text-and-visual",
+        "motionType": "guided"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Guided video generation",
+          "Controlled animation",
+          "Reference-based video creation"
+        ],
+        "platforms": ["Replicate", "FAL AI"],
+        "exampleModels": ["AnimateDiff with ControlNet"]
+      }
+    },
+    {
+      "id": "3d-to-vid",
+      "name": "3D Model to Video",
+      "input": {
+        "primary": "3d-model",
+        "secondary": []
+      },
+      "output": {
+        "primary": "video",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "rendering",
+        "renderType": "3d-rendering",
+        "motionType": "camera-path"
+      },
+      "metadata": {
+        "maturityLevel": "mature",
+        "commonUseCases": [
+          "3D visualization",
+          "Product rendering",
+          "Architectural visualization"
+        ],
+        "platforms": ["Blender", "Unreal Engine", "Unity"],
+        "exampleModels": []
+      }
+    }
+  ]
+}

taxonomy/video-generation/editing/modalities.json

@@ -0,0 +1,63 @@
+{
+  "fileType": "multimodal-ai-taxonomy",
+  "outputModality": "video",
+  "operationType": "editing",
+  "description": "Modalities for editing and transforming existing video content",
+  "modalities": [
+    {
+      "id": "vid-to-vid-no-audio",
+      "name": "Video to Video (No Audio)",
+      "input": {
+        "primary": "video",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": false
+      },
+      "characteristics": {
+        "processType": "transformation",
+        "transformationTypes": ["style-transfer", "motion-modification", "object-editing"],
+        "preserveAudio": false
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Video style transfer",
+          "Video editing",
+          "Motion manipulation"
+        ],
+        "platforms": ["Replicate", "RunwayML"],
+        "exampleModels": ["Gen-2", "Video ControlNet"]
+      }
+    },
+    {
+      "id": "vid-to-vid-preserve-audio",
+      "name": "Video to Video (Preserve Audio)",
+      "input": {
+        "primary": "video",
+        "secondary": ["text"]
+      },
+      "output": {
+        "primary": "video",
+        "audio": true,
+        "audioType": "original"
+      },
+      "characteristics": {
+        "processType": "transformation",
+        "transformationTypes": ["style-transfer", "motion-modification", "object-editing"],
+        "preserveAudio": true,
+        "audioHandling": "passthrough"
+      },
+      "metadata": {
+        "maturityLevel": "emerging",
+        "commonUseCases": [
+          "Video style transfer with audio",
+          "Content transformation maintaining soundtrack"
+        ],
+        "platforms": ["Replicate", "RunwayML"],
+        "exampleModels": []
+      }
+    }
+  ]
+}