Add lip-syncer support to source component

facefusion/filesystem.py

@@ -68,6 +68,12 @@ def is_audio(audio_path : str) -> bool:
 	return False
 
 
+def are_audios(audio_paths : List[str]) -> bool:
+	if audio_paths:
+		return all(is_audio(audio_path) for audio_path in audio_paths)
+	return False
+
+
 def is_image(image_path : str) -> bool:
 	if is_file(image_path):
 		return filetype.helpers.is_image(image_path)

facefusion/uis/components/source.py

@@ -1,35 +1,49 @@
-from typing import Optional, List
+from typing import Optional, List, Tuple
 import gradio
 
 import facefusion.globals
 from facefusion import wording
 from facefusion.uis.typing import File
-from facefusion.filesystem import are_images
+from facefusion.common_helper import get_first_item
+from facefusion.filesystem import are_audios, are_images, filter_audio_paths, filter_image_paths
 from facefusion.uis.core import register_ui_component
 
 SOURCE_FILE : Optional[gradio.File] = None
+SOURCE_AUDIO : Optional[gradio.Audio] = None
 SOURCE_IMAGE : Optional[gradio.Image] = None
 
 
 def render() -> None:
 	global SOURCE_FILE
+	global SOURCE_AUDIO
 	global SOURCE_IMAGE
 
+	are_source_audios = are_audios(facefusion.globals.source_paths)
 	are_source_images = are_images(facefusion.globals.source_paths)
 	SOURCE_FILE = gradio.File(
 		file_count = 'multiple',
 		file_types =
 		[
+			'.mp3',
+			'.wav',
 			'.png',
 			'.jpg',
 			'.webp'
 		],
 		label = wording.get('uis.source_file'),
-		value = facefusion.globals.source_paths if are_source_images else None
+		value = facefusion.globals.source_paths if are_source_audios or are_source_images else None
 	)
 	source_file_names = [ source_file_value['name'] for source_file_value in SOURCE_FILE.value ] if SOURCE_FILE.value else None
+	source_audio_path = get_first_item(filter_audio_paths(source_file_names))
+	source_image_path = get_first_item(filter_image_paths(source_file_names))
+	SOURCE_AUDIO = gradio.Audio(
+		value = source_audio_path if are_source_audios else None,
+		visible = are_source_audios,
+		show_label = False,
+		show_download_button = False
+	)
 	SOURCE_IMAGE = gradio.Image(
-		value = source_file_names[0] if are_source_images else None,
+		value = source_image_path if are_source_images else None,
 		visible = are_source_images,
 		show_label = False
 	)
@@ -37,13 +51,15 @@ def render() -> None:
 
 
 def listen() -> None:
-	SOURCE_FILE.change(update, inputs = SOURCE_FILE, outputs = SOURCE_IMAGE)
+	SOURCE_FILE.change(update, inputs = SOURCE_FILE, outputs = [ SOURCE_AUDIO, SOURCE_IMAGE ])
 
 
-def update(files : List[File]) -> gradio.Image:
+def update(files : List[File]) -> Tuple[gradio.Audio, gradio.Image]:
 	file_names = [ file.name for file in files ] if files else None
-	if are_images(file_names):
+	if are_audios(file_names) or are_images(file_names):
 		facefusion.globals.source_paths = file_names
-		return gradio.Image(value = file_names[0], visible = True)
+		audio_path = get_first_item(filter_audio_paths(facefusion.globals.source_paths))
+		image_path = get_first_item(filter_image_paths(facefusion.globals.source_paths))
+		return gradio.Audio(value = audio_path, visible = True), gradio.Image(value = image_path, visible = True)
 	facefusion.globals.source_paths = None
-	return gradio.Image(value = None, visible = False)
+	return gradio.Audio(value = None, visible = False), gradio.Image(value = None, visible = False)

requirements.txt

@@ -4,7 +4,7 @@ gradio==3.50.2
 numpy==1.26.2
 onnx==1.15.0
 onnxruntime==1.17.0
-opencv-python==4.8.1.78
+opencv-python==4.9.0.80
 psutil==5.9.6
 realesrgan==0.3.0
 torch==2.1.2

tests/test_filesystem.py

@@ -1,7 +1,7 @@
 import pytest
 
 from facefusion.download import conditional_download
-from facefusion.filesystem import is_file, is_directory, is_audio, is_image, are_images, is_video, filter_audio_paths, filter_image_paths, list_directory
+from facefusion.filesystem import is_file, is_directory, is_audio, are_audios, is_image, are_images, is_video, filter_audio_paths, filter_image_paths, list_directory
 
 
 @pytest.fixture(scope = 'module', autouse = True)
@@ -32,6 +32,12 @@ def test_is_audio() -> None:
 	assert is_audio('invalid') is False
 
 
+def test_are_audios() -> None:
+	assert are_audios([ '.assets/examples/source.mp3' ]) is True
+	assert are_audios([ '.assets/examples/source.mp3', '.assets/examples/target-240p.mp4' ]) is False
+	assert are_audios([ 'invalid' ]) is False
+
+
 def test_is_image() -> None:
 	assert is_image('.assets/examples/source.jpg') is True
 	assert is_image('.assets/examples/target-240p.mp4') is False