Fix hardcoded fps (#1068)

* fix hardcoded fps

* fallback to temp_fps when output_video_fps is None

* fix test_ffmpeg

* change temp-fps -> output-audio-fps

* cleanup

* cleanup

facefusion.ini

@@ -58,6 +58,7 @@ output_image_scale =
 output_audio_encoder =
 output_audio_quality =
 output_audio_volume =
+output_audio_fps =
 output_video_encoder =
 output_video_preset =
 output_video_quality =

facefusion/args_helper.py

@@ -52,6 +52,7 @@ def apply_args(args : Args, apply_state_item : ApplyStateItem) -> None:
 	apply_state_item('output_audio_encoder', args.get('output_audio_encoder'))
 	apply_state_item('output_audio_quality', args.get('output_audio_quality'))
 	apply_state_item('output_audio_volume', args.get('output_audio_volume'))
+	apply_state_item('output_audio_fps', normalize_fps(args.get('output_audio_fps')))
 	apply_state_item('output_video_encoder', args.get('output_video_encoder'))
 	apply_state_item('output_video_preset', args.get('output_video_preset'))
 	apply_state_item('output_video_quality', args.get('output_video_quality'))

facefusion/ffmpeg.py

@@ -235,8 +235,7 @@ def replace_audio(audio_path : str, output_path : str) -> bool:
 	return run_ffmpeg(commands).returncode == 0
 
 
-def merge_video(target_path : str, output_path : str, temp_video_fps : Fps, output_video_resolution : Resolution, trim_frame_start : int, trim_frame_end : int) -> bool:
-	output_video_fps = state_manager.get_item('output_video_fps')
+def merge_video(target_path : str, output_path : str, temp_video_fps : Fps, output_video_fps : Fps, output_video_resolution : Resolution, trim_frame_start : int, trim_frame_end : int) -> bool:
 	output_video_encoder = state_manager.get_item('output_video_encoder')
 	output_video_quality = state_manager.get_item('output_video_quality')
 	output_video_preset = state_manager.get_item('output_video_preset')

facefusion/locales.py

@@ -145,6 +145,7 @@ LOCALES : Locales =\
 			'output_audio_encoder': 'specify the encoder used for the audio',
 			'output_audio_quality': 'specify the audio quality which translates to the audio compression',
 			'output_audio_volume': 'specify the audio volume based on the target video',
+			'output_audio_fps': 'specify the fps used when converting audio to video frames',
 			'output_video_encoder': 'specify the encoder used for the video',
 			'output_video_preset': 'balance fast video processing and video file size',
 			'output_video_quality': 'specify the video quality which translates to the video compression',

facefusion/program.py

@@ -700,6 +700,17 @@ def create_output_creation_program() -> ArgumentParser:
 		],
 		scopes = [ 'api', 'cli' ]
 	)
+	capability_store.register_capability_set(
+		[
+			group_output_creation.add_argument(
+				'--output-audio-fps',
+				help = translator.get('help.output_audio_fps'),
+				type = float,
+				default = config.get_float_value('output_creation', 'output_audio_fps', '25.0')
+			)
+		],
+		scopes = [ 'api', 'cli' ]
+	)
 
 	return program
 

facefusion/types.py

@@ -435,6 +435,7 @@ StateKey = Literal\
 	'output_audio_encoder',
 	'output_audio_quality',
 	'output_audio_volume',
+	'output_audio_fps',
 	'output_video_encoder',
 	'output_video_preset',
 	'output_video_quality',
@@ -510,6 +511,7 @@ State = TypedDict('State',
 	'output_video_quality' : int,
 	'output_video_scale' : Scale,
 	'output_video_fps' : float,
+	'output_audio_fps' : float,
 	'processors' : List[str],
 	'execution_device_ids' : List[int],
 	'execution_providers' : List[ExecutionProvider],

facefusion/workflows/as_frames.py

@@ -12,13 +12,13 @@ from facefusion.workflows.core import is_process_stopping
 
 
 def create_temp_frames() -> ErrorCode:
-	state_manager.set_item('output_video_fps', 25.0)  # TODO: set default fps value
+	output_audio_fps = state_manager.get_item('output_audio_fps')
 	source_audio_path = get_first(filter_audio_paths(state_manager.get_item('source_paths')))
 	output_image_resolution = scale_resolution(detect_image_resolution(state_manager.get_item('target_path')), state_manager.get_item('output_image_scale'))
 	temp_image_resolution = restrict_image_resolution(state_manager.get_item('target_path'), output_image_resolution)
-	trim_frame_start, trim_frame_end = restrict_trim_audio_frame(source_audio_path, state_manager.get_item('output_video_fps'), state_manager.get_item('trim_frame_start'), state_manager.get_item('trim_frame_end'))
+	trim_frame_start, trim_frame_end = restrict_trim_audio_frame(source_audio_path, output_audio_fps, state_manager.get_item('trim_frame_start'), state_manager.get_item('trim_frame_end'))
 
-	if ffmpeg.spawn_frames(state_manager.get_item('target_path'), state_manager.get_item('output_path'), temp_image_resolution, state_manager.get_item('output_video_fps'), trim_frame_start, trim_frame_end):
+	if ffmpeg.spawn_frames(state_manager.get_item('target_path'), state_manager.get_item('output_path'), temp_image_resolution, output_audio_fps, trim_frame_start, trim_frame_end):
 		logger.debug(translator.get('spawning_frames_succeeded'), __name__)
 	else:
 		if is_process_stopping():

facefusion/workflows/core.py

@@ -41,11 +41,12 @@ def analyse_image() -> ErrorCode:
 def conditional_get_source_audio_frame(frame_number : int) -> AudioFrame:
 	if state_manager.get_item('workflow') in [ 'audio-to-image:frames', 'audio-to-image:video', 'image-to-video' ]:
 		source_audio_path = get_first(filter_audio_paths(state_manager.get_item('source_paths')))
-		output_video_fps = state_manager.get_item('output_video_fps')
+		fps = state_manager.get_item('output_audio_fps')
 
 		if state_manager.get_item('workflow') == 'image-to-video':
-			output_video_fps = restrict_video_fps(state_manager.get_item('target_path'), output_video_fps)
-		source_audio_frame = get_audio_frame(source_audio_path, output_video_fps, frame_number)
+			fps = restrict_video_fps(state_manager.get_item('target_path'), state_manager.get_item('output_video_fps'))
+
+		source_audio_frame = get_audio_frame(source_audio_path, fps, frame_number)
 
 		if numpy.any(source_audio_frame):
 			return source_audio_frame
@@ -56,11 +57,12 @@ def conditional_get_source_audio_frame(frame_number : int) -> AudioFrame:
 def conditional_get_source_voice_frame(frame_number: int) -> AudioFrame:
 	if state_manager.get_item('workflow') in [ 'audio-to-image:frames', 'audio-to-image:video', 'image-to-video' ]:
 		source_audio_path = get_first(filter_audio_paths(state_manager.get_item('source_paths')))
-		output_video_fps = state_manager.get_item('output_video_fps')
+		temp_fps = state_manager.get_item('output_audio_fps')
 
 		if state_manager.get_item('workflow') == 'image-to-video':
-			output_video_fps = restrict_video_fps(state_manager.get_item('target_path'), output_video_fps)
-		source_voice_frame = get_voice_frame(source_audio_path, output_video_fps, frame_number)
+			temp_fps = restrict_video_fps(state_manager.get_item('target_path'), state_manager.get_item('output_video_fps'))
+
+		source_voice_frame = get_voice_frame(source_audio_path, temp_fps, frame_number)
 
 		if numpy.any(source_voice_frame):
 			return source_voice_frame

facefusion/workflows/to_video.py

@@ -39,9 +39,10 @@ def merge_frames() -> ErrorCode:
 	trim_frame_start, trim_frame_end = restrict_trim_frame(len(temp_frame_paths), state_manager.get_item('trim_frame_start'), state_manager.get_item('trim_frame_end'))
 	output_video_resolution = conditional_scale_resolution()
 	temp_video_fps = conditional_restrict_video_fps()
+	output_fps = conditional_get_output_fps()
 
-	logger.info(translator.get('merging_video').format(resolution = pack_resolution(output_video_resolution), fps = state_manager.get_item('output_video_fps')), __name__)
-	if ffmpeg.merge_video(state_manager.get_item('target_path'), state_manager.get_item('output_path'), temp_video_fps, output_video_resolution, trim_frame_start, trim_frame_end):
+	logger.info(translator.get('merging_video').format(resolution = pack_resolution(output_video_resolution), fps = output_fps), __name__)
+	if ffmpeg.merge_video(state_manager.get_item('target_path'), state_manager.get_item('output_path'), temp_video_fps, output_fps, output_video_resolution, trim_frame_start, trim_frame_end):
 		logger.debug(translator.get('merging_video_succeeded'), __name__)
 	else:
 		if is_process_stopping():
@@ -100,6 +101,12 @@ def conditional_clear_video_pool() -> None:
 def conditional_restrict_video_fps() -> Fps:
 	if state_manager.get_item('workflow') == 'image-to-video':
 		return restrict_video_fps(state_manager.get_item('target_path'), state_manager.get_item('output_video_fps'))
+	return conditional_get_output_fps()
+
+
+def conditional_get_output_fps() -> Fps:
+	if state_manager.get_item('workflow') in [ 'audio-to-image:frames', 'audio-to-image:video' ]:
+		return state_manager.get_item('output_audio_fps')
 	return state_manager.get_item('output_video_fps')
 
 

tests/test_ffmpeg.py

@@ -132,7 +132,7 @@ def test_merge_video() -> None:
 			create_temp_directory(state_manager.get_temp_path(), output_path)
 			extract_frames(target_path, output_path, (452, 240), 25.0, 0, 1)
 
-			assert merge_video(target_path, output_path, 25.0, (452, 240), 0, 1) is True
+			assert merge_video(target_path, output_path, 25.0, 25.0, (452, 240), 0, 1) is True
 
 			clear_temp_directory(state_manager.get_temp_path(), output_path)