~

app.py

@@ -6,6 +6,7 @@ from enum import IntEnum, StrEnum, auto
 from functools import cached_property
 from pathlib import Path
 from typing import TYPE_CHECKING, Annotated, Any, TypedDict
+
 import gradio as gr
 import typer
 from huggingface_hub import snapshot_download
@@ -135,7 +136,6 @@ from ultimate_rvc.web.typing_extra import ConcurrencyId
 
 type StrPath = str | PathLike[str]
 
-
 type Json = Mapping[str, Json] | Sequence[Json] | str | int | float | bool | None
 
 
@@ -303,14 +303,12 @@ class SpeechTransferOption(StrEnum):
 class ComponentVisibilityKwArgs(TypedDict, total=False):
     """
     Keyword arguments for setting component visibility.
-
     Attributes
     ----------
     visible : bool
         Whether the component should be visible.
     value : Any
         The value of the component.
-
     """
 
     visible: bool
@@ -320,14 +318,12 @@ class ComponentVisibilityKwArgs(TypedDict, total=False):
 class UpdateDropdownKwArgs(TypedDict, total=False):
     """
     Keyword arguments for updating a dropdown component.
-
     Attributes
     ----------
     choices : DropdownChoices
         The updated choices for the dropdown component.
     value : DropdownValue
         The updated value for the dropdown component.
-
     """
 
     choices: DropdownChoices
@@ -337,14 +333,12 @@ class UpdateDropdownKwArgs(TypedDict, total=False):
 class TextBoxKwArgs(TypedDict, total=False):
     """
     Keyword arguments for updating a textbox component.
-
     Attributes
     ----------
     value : str | None
         The updated value for the textbox component.
     placeholder : str | None
         The updated placeholder for the textbox component.
-
     """
 
     value: str | None
@@ -354,12 +348,10 @@ class TextBoxKwArgs(TypedDict, total=False):
 class UpdateAudioKwArgs(TypedDict, total=False):
     """
     Keyword arguments for updating an audio component.
-
     Attributes
     ----------
     value : str | None
         The updated value for the audio component.
-
     """
 
     value: str | None
@@ -434,7 +426,6 @@ class SongIntermediateAudioConfig(BaseModel):
     """
     Configuration settings for intermediate audio components in the
     one-click song generation tab.
-
     Attributes
     ----------
     song : AudioConfig
@@ -467,7 +458,6 @@ class SongIntermediateAudioConfig(BaseModel):
         component.
     all : list[gr.Audio]
         List of instances of all intermediate audio components.
-
     """
 
     song: AudioConfig = AudioConfig.intermediate(label="Song")
@@ -505,13 +495,11 @@ class SongIntermediateAudioConfig(BaseModel):
         """
         Retrieve instances of all intermediate audio components
         in the one-click song generation tab.
-
         Returns
         -------
         list[gr.Audio]
             List of instances of all intermediate audio components in
             the one-click song generation tab.
-
         """
         # NOTE we are using self.__annotations__ to get the fields in
         # the order they are defined in the class
@@ -521,7 +509,6 @@ class SongIntermediateAudioConfig(BaseModel):
 class OneClickSongGenerationConfig(SongGenerationConfig):
     """
     Configuration settings for the one-click song generation tab.
-
     Attributes
     ----------
     n_octaves : SliderConfig
@@ -535,13 +522,11 @@ class OneClickSongGenerationConfig(SongGenerationConfig):
         component.
     intermediate_audio : SongIntermediateAudioConfig
         Configuration settings for intermediate audio components.
-
     See Also
     --------
     SongGenerationConfig
         Parent model defining common component configuration settings
         for song generation tabs.
-
     """
 
     n_octaves: SliderConfig = SliderConfig.octave_shift(
@@ -572,7 +557,6 @@ class SongInputAudioConfig(BaseModel):
     """
     Configuration settings for input audio components in the multi-step
     song generation tab.
-
     Attributes
     ----------
     audio : AudioConfig
@@ -600,7 +584,6 @@ class SongInputAudioConfig(BaseModel):
     all : list[AudioConfig]
         List of configuration settings for all input audio
         components in the multi-step song generation tab.
-
     """
 
     audio: AudioConfig = AudioConfig.input(label="Audio")
@@ -617,13 +600,11 @@ class SongInputAudioConfig(BaseModel):
         """
         Retrieve configuration settings for all input audio components
         in the multi-step song generation tab.
-
         Returns
         -------
         list[AudioConfig]
             List of configuration settings for all input audio
             components in the multi-step song generation tab.
-
         """
         return [getattr(self, field) for field in self.__annotations__]
 
@@ -632,7 +613,6 @@ class SongDirsConfig(BaseModel):
     """
     Configuration settings for song directory components in the
     multi-step song generation tab.
-
     Attributes
     ----------
     separate_audio : DropdownConfig
@@ -653,7 +633,6 @@ class SongDirsConfig(BaseModel):
     all : list[gr.Dropdown]
         List of instances of all song directory components in the
         multi-step song generation tab.
-
     """
 
     separate_audio: DropdownConfig = DropdownConfig.song_dir()
@@ -667,13 +646,11 @@ class SongDirsConfig(BaseModel):
         """
         Retrieve instances of all song directory components in the
         multi-step song generation tab.
-
         Returns
         -------
         list[gr.Dropdown]
             List of instances of all song directory components in
             the multi-step song generation tab.
-
         """
         return [getattr(self, field).instance for field in self.__annotations__]
 
@@ -681,7 +658,6 @@ class SongDirsConfig(BaseModel):
 class MultiStepSongGenerationConfig(SongGenerationConfig):
     """
     Configuration settings for multi-step song generation tab.
-
     Attributes
     ----------
     separation_model : DropdownConfig
@@ -705,13 +681,11 @@ class MultiStepSongGenerationConfig(SongGenerationConfig):
         Configuration settings for input audio components.
     song_dirs : SongDirsConfig
         Configuration settings for song directory components.
-
     See Also
     --------
     SongGenerationConfig
         Parent model defining common component configuration settings
         for song generation tabs.
-
     """
 
     separation_model: DropdownConfig = DropdownConfig(
@@ -759,7 +733,6 @@ class SpeechIntermediateAudioConfig(BaseModel):
     """
     Configuration settings for intermediate audio components in the
     one-click speech generation tab.
-
     Attributes
     ----------
     speech : AudioConfig
@@ -769,7 +742,6 @@ class SpeechIntermediateAudioConfig(BaseModel):
     all : list[gr.Audio]
         List of instances of all intermediate audio components in the
         speech generation tab.
-
     """
 
     speech: AudioConfig = AudioConfig.intermediate(label="Speech")
@@ -780,13 +752,11 @@ class SpeechIntermediateAudioConfig(BaseModel):
         """
         Retrieve instances of all intermediate audio components in the
         speech generation tab.
-
         Returns
         -------
         list[gr.Audio]
             List of instances of all intermediate audio components in
             the speech generation tab.
-
         """
         return [getattr(self, field).instance for field in self.__annotations__]
 
@@ -794,7 +764,6 @@ class SpeechIntermediateAudioConfig(BaseModel):
 class OneClickSpeechGenerationConfig(SpeechGenerationConfig):
     """
     Configuration settings for one-click speech generation tab.
-
     Attributes
     ----------
     intermediate_audio : SpeechIntermediateAudioConfig
@@ -802,13 +771,11 @@ class OneClickSpeechGenerationConfig(SpeechGenerationConfig):
     show_intermediate_audio : CheckboxConfig
         Configuration settings for a show intermediate audio checkbox
         component.
-
     See Also
     --------
     SpeechGenerationConfig
         Parent model defining common component configuration settings
         for speech generation tabs.
-
     """
 
     intermediate_audio: SpeechIntermediateAudioConfig = SpeechIntermediateAudioConfig()
@@ -825,18 +792,15 @@ class SpeechInputAudioConfig(BaseModel):
     """
     Configuration settings for input audio components in the multi-step
     speech generation tab.
-
     Attributes
     ----------
     speech : AudioConfig
         Configuration settings for the input speech audio component.
     converted_speech : AudioConfig
         Configuration settings for the converted speech audio component.
-
     all : list[AudioConfig]
         List of configuration settings for all input audio components in
         the multi-step speech generation tab.
-
     """
 
     speech: AudioConfig = AudioConfig.input("Speech")
@@ -847,13 +811,11 @@ class SpeechInputAudioConfig(BaseModel):
         """
         Retrieve configuration settings for all input audio components
         in the multi-step speech generation tab.
-
         Returns
         -------
         list[AudioConfig]
             List of configuration settings for all input audio
             components in the multi-step speech generation tab.
-
         """
         return [getattr(self, field) for field in self.__annotations__]
 
@@ -861,18 +823,15 @@ class SpeechInputAudioConfig(BaseModel):
 class MultiStepSpeechGenerationConfig(SpeechGenerationConfig):
     """
     Configuration settings for the multi-step speech generation tab.
-
     Attributes
     ----------
     input_audio : SpeechInputAudioConfig
         Configuration settings for input audio components.
-
     See Also
     --------
     SpeechGenerationConfig
         Parent model defining common component configuration settings
         for speech generation tabs.
-
     """
 
     input_audio: SpeechInputAudioConfig = SpeechInputAudioConfig()
@@ -884,9 +843,7 @@ class MultiStepTrainingConfig(TrainingConfig):
 
 class ModelManagementConfig(BaseModel):
     """
-
     Configuration settings for model management tab.
-
     Attributes
     ----------
     voices : DropdownConfig
@@ -903,7 +860,6 @@ class ModelManagementConfig(BaseModel):
         component.
     dummy_checkbox : CheckboxConfig
         Configuration settings for a dummy checkbox component.
-
     """
 
     voices: DropdownConfig = DropdownConfig.multi_delete(
@@ -933,7 +889,6 @@ class ModelManagementConfig(BaseModel):
 class AudioManagementConfig(BaseModel):
     """
     Configuration settings for audio management tab.
-
     Attributes
     ----------
     intermediate : DropdownConfig
@@ -950,7 +905,6 @@ class AudioManagementConfig(BaseModel):
         component.
     dummy_checkbox : CheckboxConfig
         Configuration settings for a dummy checkbox component.
-
     """
 
     intermediate: DropdownConfig = DropdownConfig.multi_delete(
@@ -983,12 +937,10 @@ class AudioManagementConfig(BaseModel):
 class SettingsManagementConfig(BaseModel):
     """
     Configuration settings for settings management tab.
-
     Attributes
     ----------
     dummy_checkbox : CheckboxConfig
         Configuration settings for a dummy checkbox component.
-
     """
 
     load_config_name: DropdownConfig = DropdownConfig(
@@ -1012,14 +964,12 @@ class SettingsManagementConfig(BaseModel):
 class TotalSongGenerationConfig(BaseModel):
     """
     All configuration settings for song generation tabs.
-
     Attributes
     ----------
     one_click : OneClickSongGenerationConfig
         Configuration settings for the one-click song generation tab.
     multi_step : MultiStepSongGenerationConfig
         Configuration settings for the multi-step song generation tab.
-
     """
 
     one_click: OneClickSongGenerationConfig = OneClickSongGenerationConfig()
@@ -1029,14 +979,12 @@ class TotalSongGenerationConfig(BaseModel):
 class TotalSpeechGenerationConfig(BaseModel):
     """
     All configuration settings for speech generation tabs.
-
     Attributes
     ----------
     one_click : OneClickSpeechGenerationConfig
         Configuration settings for the one-click speech generation tab.
     multi_step : MultiStepSpeechGenerationConfig
         Configuration settings for the multi-step speech generation tab.
-
     """
 
     one_click: OneClickSpeechGenerationConfig = OneClickSpeechGenerationConfig()
@@ -1046,12 +994,10 @@ class TotalSpeechGenerationConfig(BaseModel):
 class TotalTrainingConfig(BaseModel):
     """
     All configuration settings for training tabs.
-
     Attributes
     ----------
     training : TrainingConfig
         Configuration settings for the multi-step training tab.
-
     """
 
     multi_step: MultiStepTrainingConfig = MultiStepTrainingConfig()
@@ -1060,7 +1006,6 @@ class TotalTrainingConfig(BaseModel):
 class TotalManagementConfig(BaseModel):
     """
     All configuration settings for management tabs.
-
     Attributes
     ----------
     model : ModelManagementConfig
@@ -1069,7 +1014,6 @@ class TotalManagementConfig(BaseModel):
         Configuration settings for the audio management tab.
     settings : SettingsManagementConfig
         Configuration settings for the settings management tab.
-
     """
 
     model: ModelManagementConfig = ModelManagementConfig()
@@ -1080,7 +1024,6 @@ class TotalManagementConfig(BaseModel):
 class TotalConfig(BaseModel):
     """
     All configuration settings for the Ultimate RVC app.
-
     Attributes
     ----------
     song : TotalSongGenerationConfig
@@ -1091,7 +1034,6 @@ class TotalConfig(BaseModel):
         Configuration settings for training tabs.
     management : TotalManagementConfig
         Configuration settings for management tabs.
-
     """
 
     song: TotalSongGenerationConfig = TotalSongGenerationConfig()
@@ -1105,14 +1047,12 @@ class TotalConfig(BaseModel):
         Recursively collect those component configuration models nested
         within the current model instance, which have values that are
         not excluded.
-
         Returns
         -------
         list[AnyComponentConfig]
             A list of component configuration models found within the
             current model instance, which have values that are not
             excluded.
-
         """
 
         def _collect(model: BaseModel) -> list[AnyComponentConfig]:
@@ -1132,7 +1072,6 @@ class BaseTabConfig(BaseModel):
     """
     Base model defining common component configuration settings for
     UI tabs.
-
     Attributes
     ----------
     embedder_model : DropdownConfig
@@ -1140,7 +1079,6 @@ class BaseTabConfig(BaseModel):
     custom_embedder_model : DropdownConfig
         Configuration settings for a custom embedder model dropdown
         component.
-
     """
 
     embedder_model: DropdownConfig = DropdownConfig(
@@ -1163,7 +1101,6 @@ class BaseTabConfig(BaseModel):
 class TrainingConfig(BaseTabConfig):
     """
     Common component configuration settings for training tabs.
-
     Attributes
     ----------
     dataset_type : DropdownConfig
@@ -1262,13 +1199,11 @@ class TrainingConfig(BaseTabConfig):
     reduce_memory_usage : CheckboxConfig
         Configuration settings for a reduce-memory-usage checkbox
         component.
-
     See Also
     --------
     BaseTabConfig
         Parent model defining common component configuration settings
         for UI tabs.
-
     """
 
     dataset_type: DropdownConfig = DropdownConfig(
@@ -1386,7 +1321,7 @@ class TrainingConfig(BaseTabConfig):
         exclude_value=True,
     )
 
-    hop_length: SliderConfig = SliderConfig(
+    hop_length: SliderConfig = SliderConfig.hop_length(
         label="Hop length",
         info="The hop length to use for extracting pitch features.<br><br>",
         visible=False,
@@ -1578,7 +1513,6 @@ class TrainingConfig(BaseTabConfig):
 class GenerationConfig(BaseTabConfig):
     """
     Common component configuration settings for generation tabs.
-
     voice_model : DropdownConfig
         Configuration settings for a voice model dropdown component.
     f0_methods : DropdownConfig
@@ -1607,13 +1541,11 @@ class GenerationConfig(BaseTabConfig):
         component.
     output_name : TextboxConfig
         Configuration settings for an output name textbox component.
-
     See Also
     --------
     BaseTabConfig
         Parent model defining common component configuration settings
         for UI tabs.
-
     """
 
     voice_model: DropdownConfig = DropdownConfig(
@@ -1732,7 +1664,6 @@ class GenerationConfig(BaseTabConfig):
 class SongGenerationConfig(GenerationConfig):
     """
     Common component configuration settings for song generation tabs.
-
     Attributes
     ----------
     source_type : DropdownConfig
@@ -1761,13 +1692,11 @@ class SongGenerationConfig(GenerationConfig):
     backup_gain : SliderConfig
         Configuration settings for a backup vocals gain slider
         component.
-
     See Also
     --------
     GenerationConfig
         Parent model defining common component configuration settings
         for song generation tabs.
-
     """
 
     source_type: DropdownConfig = DropdownConfig(
@@ -1850,7 +1779,6 @@ class SongGenerationConfig(GenerationConfig):
 class SpeechGenerationConfig(GenerationConfig):
     """
     Common component configuration settings for speech generation tabs.
-
     Attributes
     ----------
     source_type : DropdownConfig
@@ -1883,13 +1811,11 @@ class SpeechGenerationConfig(GenerationConfig):
         component.
     output_gain : GainSliderConfig
         Configuration settings for an output gain slider component.
-
     See Also
     --------
     GenerationConfig
         Parent model defining common component configuration settings
         for generation tabs.
-
     """
 
     source_type: DropdownConfig = DropdownConfig(
@@ -1975,7 +1901,6 @@ def render_song_cover_multi_step_tab(
 ) -> None:
     """
     Render "Generate song cover - multi-step generation" tab.
-
     Parameters
     ----------
     total_config : TotalConfig
@@ -1984,7 +1909,6 @@ def render_song_cover_multi_step_tab(
     cookiefile : str, optional
         The path to a file containing cookies to use when downloading
         audio from Youtube.
-
     """
     tab_config = total_config.song.multi_step
     for input_track in tab_config.input_audio.all:
@@ -2618,7 +2542,6 @@ def _pair_audio_tracks_and_gain(
     """
     Pair audio tracks and gain levels stored in separate gradio
     components.
-
     This function is meant to first be partially applied to the sequence
     of audio components and the sequence of slider components containing
     the values that should be combined. The resulting function can then
@@ -2626,7 +2549,6 @@ def _pair_audio_tracks_and_gain(
     those audio and slider components. The `data` parameter in that case
     will contain a mapping from each of those components to the value
     that the component stores.
-
     Parameters
     ----------
     audio_components : Sequence[gr.Audio]
@@ -2635,17 +2557,14 @@ def _pair_audio_tracks_and_gain(
         Gain level components to pair with audio tracks.
     data : dict[gr.Audio | gr.Slider, Any]
         Data from the audio and gain components.
-
     Returns
     -------
     list[tuple[str, int]]
         Paired audio tracks and gain levels.
-
     Raises
     ------
     ValueError
         If the number of audio tracks and gain levels are not the same.
-
     """
     audio_tracks = [data[component] for component in audio_components]
     gain_levels = [data[component] for component in gain_components]
@@ -2659,19 +2578,244 @@ def _pair_audio_tracks_and_gain(
     ]
 
 
+def run_newpipeline(
+    source: str,
+    model_name: str,
+    n_octaves: int = 0,
+    n_semitones: int = 0,
+    f0_methods: Sequence[F0Method] | None = None,
+    index_rate: float = 0.3,
+    rms_mix_rate: float = 1.0,
+    protect_rate: float = 0.33,
+    hop_length: int = 128,
+    split_vocals: bool = False,
+    autotune_vocals: bool = False,
+    autotune_strength: float = 1.0,
+    clean_vocals: bool = False,
+    clean_strength: float = 0.7,
+    embedder_model: EmbedderModel = EmbedderModel.CONTENTVEC,
+    custom_embedder_model: str | None = None,
+    sid: int = 0,
+    room_size: float = 0.15,
+    wet_level: float = 0.2,
+    dry_level: float = 0.8,
+    damping: float = 0.7,
+    main_gain: int = 0,
+    inst_gain: int = 0,
+    backup_gain: int = 0,
+    output_sr: int = 44100,
+    output_format: AudioExt = AudioExt.MP3,
+    output_name: str | None = None,
+    cookiefile: StrPath | None = None,
+    progress_bar: gr.Progress | None = None,
+) -> tuple[Path, ...]:
+    """
+    Run the song cover generation pipeline.
+    Parameters
+    ----------
+    source : str
+        A Youtube URL, the path to a local audio file or the path to a
+        song directory.
+    model_name : str
+        The name of the voice model to use for vocal conversion.
+    n_octaves : int, default=0
+        The number of octaves to pitch-shift the converted vocals by.
+    n_semitones : int, default=0
+        The number of semi-tones to pitch-shift the converted vocals,
+        instrumentals, and backup vocals by.
+    f0_methods : Sequence[F0Method], optional
+        The methods to use for pitch extraction during vocal
+        conversion. If None, the method used is rmvpe.
+    index_rate : float, default=0.3
+        The influence of the index file on the vocal conversion.
+    rms_mix_rate : float, default=1.0
+        The blending rate of the volume envelope of the converted
+        vocals.
+    protect_rate : float, default=0.33
+        The protect rate for consonants and breathing sounds during
+        vocal conversion.
+    hop_length : int, default=128
+        The hop length to use for crepe-based pitch detection.
+    split_vocals : bool, default=False
+        Whether to perform audio splitting before converting the main
+        vocals.
+    autotune_vocals : bool, default=False
+        Whether to apply autotune to the converted vocals.
+    autotune_strength : float, default=1.0
+        The strength of the autotune to apply to the converted vocals.
+    clean_vocals : bool, default=False
+        Whether to clean the converted vocals.
+    clean_strength : float, default=0.7
+        The intensity of the cleaning to apply to the converted vocals.
+    embedder_model : EmbedderModel, default=EmbedderModel.CONTENTVEC
+        The model to use for generating speaker embeddings during vocal
+        conversion.
+    custom_embedder_model : StrPath, optional
+        The name of a custom embedder model to use for generating
+        speaker embeddings during vocal conversion.
+    sid : int, default=0
+        The speaker id to use for multi-speaker models during vocal
+        conversion.
+    room_size : float, default=0.15
+        The room size of the reverb effect to apply to the converted
+        vocals.
+    wet_level : float, default=0.2
+        The wetness level of the reverb effect to apply to the converted
+        vocals.
+    dry_level : float, default=0.8
+        The dryness level of the reverb effect to apply to the converted
+        vocals.
+    damping : float, default=0.7
+        The damping of the reverb effect to apply to the converted
+        vocals.
+    main_gain : int, default=0
+        The gain to apply to the post-processed vocals.
+    inst_gain : int, default=0
+        The gain to apply to the pitch-shifted instrumentals.
+    backup_gain : int, default=0
+        The gain to apply to the pitch-shifted backup vocals.
+    output_sr : int, default=44100
+        The sample rate of the song cover.
+    output_format : AudioExt, default=AudioExt.MP3
+        The audio format of the song cover.
+    output_name : str, optional
+        The name of the song cover.
+    cookiefile : StrPath, optional
+        The path to a file containing cookies to use when downloading
+        audio from Youtube.
+    progress_bar : gr.Progress, optional
+        Gradio progress bar to update.
+    Returns
+    -------
+    tuple[Path,...]
+        The path to the generated song cover and the paths to any
+        intermediate audio files that were generated.
+    """
+    validate_model(model_name, Entity.VOICE_MODEL)
+    if embedder_model == EmbedderModel.CUSTOM:
+        validate_model(custom_embedder_model, Entity.CUSTOM_EMBEDDER_MODEL)
+    display_progress("[~] Retrieving song...", 0 / 9, progress_bar)
+    song, song_dir = retrieve_song(source, cookiefile=cookiefile)
+    display_progress(
+        "[~] newpipeline: Separating vocals from instrumentals...", 1 / 9, progress_bar
+    )
+    vocals_track, instrumentals_track = separate_audio(
+        song,
+        song_dir,
+        SeparationModel.MDX23C_8KFFT_InstVoc_HQ_2,
+        SegmentSize.SEG_2048,
+    )
+    display_progress(
+        "[~] newpipeline: Separating main vocals from backup vocals...",
+        2 / 9,
+        progress_bar,
+    )
+    backup_vocals_track, main_vocals_track = separate_audio(
+        vocals_track,
+        song_dir,
+        SeparationModel.UVR_MDX_NET_KARA_2,
+        SegmentSize.SEG_2048,
+    )
+    display_progress("[~] newpipeline: De-noising vocals...", 3 / 9, progress_bar)
+    noise_track, clean_track = separate_audio(
+        clean_track,
+        song_dir,
+        SeparationModel.UVR_DeNoise,
+        SegmentSize.SEG_2048,
+    )
+    display_progress("[~] newpipeline: De-reverbing vocals...", 4 / 9, progress_bar)
+    reverb_track, vocals_dereverb_track = separate_audio(
+        main_vocals_track,
+        song_dir,
+        SeparationModel.UVR_DeEcho_DeReverb,
+        SegmentSize.SEG_2048,
+    )
+
+    display_progress("[~] newpipeline: Converting vocals...", 5 / 9, progress_bar)
+    converted_vocals_track = convert(
+        audio_track=vocals_dereverb_track,
+        directory=song_dir,
+        model_name=model_name,
+        n_octaves=n_octaves,
+        n_semitones=n_semitones,
+        f0_methods=f0_methods,
+        index_rate=index_rate,
+        rms_mix_rate=rms_mix_rate,
+        protect_rate=protect_rate,
+        hop_length=hop_length,
+        split_audio=split_vocals,
+        autotune_audio=autotune_vocals,
+        autotune_strength=autotune_strength,
+        clean_audio=clean_vocals,
+        clean_strength=clean_strength,
+        embedder_model=embedder_model,
+        custom_embedder_model=custom_embedder_model,
+        sid=sid,
+        content_type=RVCContentType.VOCALS,
+    )
+    display_progress("[~] newpipeline: Post-processing vocals...", 6 / 9, progress_bar)
+    effected_vocals_track = postprocess(
+        converted_vocals_track,
+        song_dir,
+        room_size,
+        wet_level,
+        dry_level,
+        damping,
+    )
+    display_progress(
+        "[~] newpipeline: Pitch-shifting instrumentals...", 7 / 9, progress_bar
+    )
+    shifted_instrumentals_track = pitch_shift(
+        instrumentals_track,
+        song_dir,
+        n_semitones,
+    )
+    display_progress(
+        "[~] newpipeline: Pitch-shifting backup vocals...", 8 / 9, progress_bar
+    )
+    shifted_backup_vocals_track = pitch_shift(
+        backup_vocals_track,
+        song_dir,
+        n_semitones,
+    )
+
+    song_cover = mix_song(
+        [
+            (effected_vocals_track, main_gain),
+            (shifted_instrumentals_track, inst_gain),
+            (shifted_backup_vocals_track, backup_gain),
+        ],
+        song_dir,
+        output_sr,
+        output_format,
+        output_name,
+    )
+    return (
+        song_cover,
+        song,
+        vocals_track,
+        instrumentals_track,
+        main_vocals_track,
+        backup_vocals_track,
+        vocals_dereverb_track,
+        reverb_track,
+        converted_vocals_track,
+        effected_vocals_track,
+        shifted_instrumentals_track,
+        shifted_backup_vocals_track,
+    )
+
+
 def render_app() -> gr.Blocks:
     """
     Render the Ultimate RVC web application.
-
     Returns
     -------
     gr.Blocks
         The rendered web application.
-
     """
     css = """
     h1 { text-align: center; margin-top: 20px; margin-bottom: 20px; }
-
     #generate-tab-button { font-weight: bold !important;}
     #manage-tab-button { font-weight: bold !important;}
     #audio-tab-button { font-weight: bold !important;}
@@ -2783,13 +2927,11 @@ def _init_dropdowns() -> list[gr.Dropdown]:
     """
     Initialize the Ultimate RVC web application by updating the choices
     and default values of non-static dropdown components.
-
     Returns
     -------
     tuple[gr.Dropdown, ...]
         A tuple of gr.Dropdown components with updated choices and
         default values.
-
     """
     # Initialize model dropdowns
     edge_tts_models = initialize_dropdowns(
@@ -2848,7 +2990,6 @@ def render_song_cover_one_click_tab(
 ) -> None:
     """
     Render "Generate song covers - One-click generation" tab.
-
     Parameters
     ----------
     total_config : TotalConfig
@@ -2857,7 +2998,6 @@ def render_song_cover_one_click_tab(
     cookiefile : str, optional
         The path to a file containing cookies to use when downloading
         audio from Youtube.
-
     """
     with gr.Tab("One-click"):
         tab_config = total_config.song.one_click
@@ -3246,7 +3386,7 @@ def start_app(
     app.launch(
         server_name=listen_host,
         server_port=listen_port,
-        ssr_mode=ssr_mode,
+        ssr_mode=false,
     )