### Run Local Gradio Demo for FantasyTalking

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Instructions to install necessary Python packages ('gradio', 'spaces') and run the local Gradio demonstration application for FantasyTalking.

```Bash
pip install gradio spaces
python app.py
```

--------------------------------

### Clone and Install FantasyTalking Dependencies

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Instructions to clone the FantasyTalking repository and install its required Python dependencies. It also mentions an optional step to install 'flash_attn' for accelerated attention computation.

```Bash
git clone https://github.com/Fantasy-AMAP/fantasy-talking.git
cd fantasy-talking
```

```Bash
# Ensure torch >= 2.0.0
pip install -r requirements.txt
# Optional to install flash_attn to accelerate attention computation
pip install flash_attn
```

--------------------------------

### Download FantasyTalking Models using Huggingface CLI

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Commands to install the Huggingface CLI and download the necessary models (Wan2.1-I2V-14B-720P, wav2vec2-base-960h, and FantasyTalking checkpoint) to local directories.

```Bash
pip install "huggingface_hub[cli]"
huggingface-cli download Wan-AI/Wan2.1-I2V-14B-720P --local-dir ./models/Wan2.1-I2V-14B-720P
huggingface-cli download facebook/wav2vec2-base-960h --local-dir ./models/wav2vec2-base-960h
huggingface-cli download acvlab/FantasyTalking fantasytalking_model.ckpt --local-dir ./models
```

--------------------------------

### Download FantasyTalking Models using ModelScope CLI

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Commands to install the ModelScope CLI and download the necessary models (Wan2.1-I2V-14B-720P, wav2vec2-base-960h, and FantasyTalking checkpoint) to local directories.

```Bash
pip install modelscope
modelscope download Wan-AI/Wan2.1-I2V-14B-720P --local_dir ./models/Wan2.1-I2V-14B-720P
modelscope download AI-ModelScope/wav2vec2-base-960h --local_dir ./models/wav2vec2-base-960h
modelscope download amap_cvlab/FantasyTalking   fantasytalking_model.ckpt  --local_dir ./models
```

--------------------------------

### List Python Project Dependencies

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/requirements.txt

This snippet enumerates the Python packages and their exact versions necessary for the '/fantasy-amap/fantasy-talking' project. These dependencies are typically found in a `requirements.txt` file and are essential for setting up the project's runtime environment, ensuring all required libraries are installed and compatible.

```Python
torch>=2.0.0
torchvision
cupy-cuda12x
transformers==4.46.2
controlnet-aux==0.0.7
imageio
imageio[ffmpeg]
safetensors
einops
sentencepiece
protobuf
modelscope
ftfy
librosa
```

--------------------------------

### Run Basic FantasyTalking Inference

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Command to run the 'infer.py' script with a specified image and audio path to generate a talking portrait.

```Bash
python infer.py  --image_path ./assets/images/woman.png --audio_path ./assets/audios/woman.wav
```

--------------------------------

### Run FantasyTalking Inference with Prompt and CFG Scale

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

Command to run the 'infer.py' script, allowing control over the character's behavior using a text prompt and adjusting the prompt and audio CFG scales for more consistent lip-sync.

```Bash
python infer.py  --image_path ./assets/images/woman.png --audio_path ./assets/audios/woman.wav --prompt "The person is speaking enthusiastically, with their hands continuously waving." --prompt_cfg_scale 5.0 --audio_cfg_scale 5.0
```

--------------------------------

### Cite FantasyTalking Project

Source: https://github.com/fantasy-amap/fantasy-talking/blob/main/README.md

BibTeX entry for citing the FantasyTalking project, useful for academic papers and research.

```BibTeX
@article{wang2025fantasytalking,
   title={FantasyTalking: Realistic Talking Portrait Generation via Coherent Motion Synthesis},
   author={Wang, Mengchao and Wang, Qiang and Jiang, Fan and Fan, Yaqi and Zhang, Yunpeng and Qi, Yonggang and Zhao, Kun and Xu, Mu},
   journal={arXiv preprint arXiv:2504.04842},
   year={2025}
 }
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.