Update to use current version of libraries

gradio_app.py

@@ -158,4 +158,4 @@ def create_args(
     )
 
 if __name__ == "__main__":
-    demo.launch(inbrowser=True, share=True)
+    demo.launch(inbrowser=True, share=False)

latentsync/models/attention.py

@@ -9,10 +9,10 @@
 from torch import nn
 
 from diffusers.configuration_utils import ConfigMixin, register_to_config
-from diffusers.modeling_utils import ModelMixin
+from diffusers.models.modeling_utils import ModelMixin
 from diffusers.utils import BaseOutput
 from diffusers.utils.import_utils import is_xformers_available
-from diffusers.models.attention import CrossAttention, FeedForward, AdaLayerNorm
+from diffusers.models.attention import Attention as CrossAttention, FeedForward, AdaLayerNorm
 
 from einops import rearrange, repeat
 from .utils import zero_module

latentsync/models/motion_module.py

@@ -11,10 +11,10 @@
 from torch import nn
 
 from diffusers.configuration_utils import ConfigMixin, register_to_config
-from diffusers.modeling_utils import ModelMixin
+from diffusers.models.modeling_utils import ModelMixin
 from diffusers.utils import BaseOutput
 from diffusers.utils.import_utils import is_xformers_available
-from diffusers.models.attention import CrossAttention, FeedForward
+from diffusers.models.attention import Attention as CrossAttention, FeedForward
 
 from einops import rearrange, repeat
 import math

latentsync/models/unet.py

@@ -9,7 +9,7 @@
 import torch.utils.checkpoint
 
 from diffusers.configuration_utils import ConfigMixin, register_to_config
-from diffusers.modeling_utils import ModelMixin
+from diffusers.models.modeling_utils import ModelMixin
 from diffusers import UNet2DConditionModel
 from diffusers.utils import BaseOutput, logging
 from diffusers.models.embeddings import TimestepEmbedding, Timesteps

latentsync/pipelines/lipsync_pipeline.py

@@ -15,7 +15,7 @@
 
 from diffusers.configuration_utils import FrozenDict
 from diffusers.models import AutoencoderKL
-from diffusers.pipeline_utils import DiffusionPipeline
+from diffusers import DiffusionPipeline
 from diffusers.schedulers import (
     DDIMScheduler,
     DPMSolverMultistepScheduler,

requirements.txt

@@ -1,30 +1,30 @@
-torch==2.2.2
-torchvision==0.17.2
---extra-index-url https://download.pytorch.org/whl/cu121
-xformers==0.0.26
-triton==2.2.0
+torch
+torchvision
+--extra-index-url https://download.pytorch.org/whl/cu124
+xformers
+triton
 
-diffusers==0.11.1
-transformers==4.38.0
-huggingface-hub==0.25.2
-imageio==2.27.0
-decord==0.6.0
-accelerate==0.26.1
-einops==0.7.0
-omegaconf==2.3.0
-safetensors==0.4.2
-opencv-python==4.9.0.80
-mediapipe==0.10.11
-av==11.0.0
-torch-fidelity==0.3.0
-torchmetrics==1.3.1
-python_speech_features==0.6
-librosa==0.10.1
-scenedetect==0.6.1
-ffmpeg-python==0.2.0
-lpips==0.1.4
-face-alignment==1.4.1
-ninja==1.11.1.1
-pandas==2.0.3
+diffusers
+transformers
+huggingface-hub
+imageio
+decord
+accelerate
+einops
+omegaconf
+safetensors
+opencv-python
+mediapipe
+av
+torch-fidelity
+torchmetrics
+python_speech_features
+librosa
+scenedetect
+ffmpeg-python
+lpips
+face-alignment
+ninja
+pandas
 numpy==1.24.4
-gradio==5.9.1
+gradio

scripts/inference.py

@@ -52,8 +52,8 @@ def main(config, args):
     unet = unet.to(dtype=torch.float16)
 
     # set xformers
-    if is_xformers_available():
-        unet.enable_xformers_memory_efficient_attention()
+#    if is_xformers_available():
+#        unet.enable_xformers_memory_efficient_attention()
 
     pipeline = LipsyncPipeline(
         vae=vae,

setup_env.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 
 # Create a new conda environment
-conda create -y -n latentsync python=3.10.13
+conda create -y -n latentsync python=3.11
 conda activate latentsync
 
 # Install ffmpeg
@@ -20,4 +20,4 @@ huggingface-cli download chunyu-li/LatentSync --local-dir checkpoints --exclude
 mkdir -p ~/.cache/torch/hub/checkpoints
 ln -s $(pwd)/checkpoints/auxiliary/2DFAN4-cd938726ad.zip ~/.cache/torch/hub/checkpoints/2DFAN4-cd938726ad.zip
 ln -s $(pwd)/checkpoints/auxiliary/s3fd-619a316812.pth ~/.cache/torch/hub/checkpoints/s3fd-619a316812.pth
-ln -s $(pwd)/checkpoints/auxiliary/vgg16-397923af.pth ~/.cache/torch/hub/checkpoints/vgg16-397923af.pth
+ln -s $(pwd)/checkpoints/auxiliary/vgg16-397923af.pth ~/.cache/torch/hub/checkpoints/vgg16-397923af.pth