app.py

from typing import Optional, Any
import os
import sys
import torch
import logging
import yt_dlp
from yt_dlp import YoutubeDL
import gradio as gr
import argparse
from audio_separator.separator import Separator
import numpy as np
import librosa
import soundfile as sf
from ensemble import ensemble_files
import shutil
import gradio_client.utils as client_utils
import matchering as mg
import spaces
import gdown
from pydub import AudioSegment
import gc
import time
from concurrent.futures import ThreadPoolExecutor, as_completed
from threading import Lock
import scipy.io.wavfile

# Logging setup
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Gradio JSON schema patch
original_json_schema_to_python_type = client_utils._json_schema_to_python_type

def patched_json_schema_to_python_type(schema: Any, defs: Optional[dict] = None) -> str:
    logger.debug(f"Parsing schema: {schema}")
    if isinstance(schema, bool):
        logger.info("Found boolean schema, returning 'boolean'")
        return "boolean"
    if not isinstance(schema, dict):
        logger.warning(f"Unexpected schema type: {type(schema)}, returning 'Any'")
        return "Any"
    if "enum" in schema and schema.get("type") == "string":
        logger.info(f"Handling enum schema: {schema['enum']}")
        return f"Literal[{', '.join(repr(e) for e in schema['enum'])}]"
    try:
        return original_json_schema_to_python_type(schema, defs)
    except client_utils.APIInfoParseError as e:
        logger.error(f"Failed to parse schema {schema}: {e}")
        return "str"

client_utils._json_schema_to_python_type = patched_json_schema_to_python_type

# Device setup
device = "cuda" if torch.cuda.is_available() else "cpu"
use_autocast = device == "cuda"
logger.info(f"Using device: {device}")

# ROFORMER_MODELS and OUTPUT_FORMATS
ROFORMER_MODELS = {
    "Vocals": {
        'MelBand Roformer | Big Beta 6X by unwa': 'melband_roformer_big_beta6x.ckpt',
        'MelBand Roformer Kim | Big Beta 4 FT by unwa': 'melband_roformer_big_beta4.ckpt',
        'MelBand Roformer Kim | Big Beta 5e FT by unwa': 'melband_roformer_big_beta5e.ckpt',
        'MelBand Roformer | Big Beta 6 by unwa': 'melband_roformer_big_beta6.ckpt',
        'MelBand Roformer | Vocals by Kimberley Jensen': 'vocals_mel_band_roformer.ckpt',
        'MelBand Roformer Kim | FT 3 by unwa': 'mel_band_roformer_kim_ft3_unwa.ckpt',
        'MelBand Roformer Kim | FT by unwa': 'mel_band_roformer_kim_ft_unwa.ckpt',
        'MelBand Roformer Kim | FT 2 by unwa': 'mel_band_roformer_kim_ft2_unwa.ckpt',
        'MelBand Roformer Kim | FT 2 Bleedless by unwa': 'mel_band_roformer_kim_ft2_bleedless_unwa.ckpt',
        'MelBand Roformer | Vocals by becruily': 'mel_band_roformer_vocals_becruily.ckpt',
        'MelBand Roformer | Vocals Fullness by Aname': 'mel_band_roformer_vocal_fullness_aname.ckpt',
        'BS Roformer | Vocals by Gabox': 'bs_roformer_vocals_gabox.ckpt',
        'MelBand Roformer | Vocals by Gabox': 'mel_band_roformer_vocals_gabox.ckpt',
        'MelBand Roformer | Vocals FV1 by Gabox': 'mel_band_roformer_vocals_fv1_gabox.ckpt',
        'MelBand Roformer | Vocals FV2 by Gabox': 'mel_band_roformer_vocals_fv2_gabox.ckpt',
        'MelBand Roformer | Vocals FV3 by Gabox': 'mel_band_roformer_vocals_fv3_gabox.ckpt',
        'MelBand Roformer | Vocals FV4 by Gabox': 'mel_band_roformer_vocals_fv4_gabox.ckpt',
        'BS Roformer | Chorus Male-Female by Sucial': 'model_chorus_bs_roformer_ep_267_sdr_24.1275.ckpt',
        'BS Roformer | Male-Female by aufr33': 'bs_roformer_male_female_by_aufr33_sdr_7.2889.ckpt',
    },
    "Instrumentals": {
        'MelBand Roformer | FVX by Gabox': 'mel_band_roformer_instrumental_fvx_gabox.ckpt',
        'MelBand Roformer | INSTV8N by Gabox': 'mel_band_roformer_instrumental_instv8n_gabox.ckpt',
        'MelBand Roformer | INSTV8 by Gabox': 'mel_band_roformer_instrumental_instv8_gabox.ckpt',
        'MelBand Roformer | INSTV7N by Gabox': 'mel_band_roformer_instrumental_instv7n_gabox.ckpt',
        'MelBand Roformer | Instrumental Bleedless V3 by Gabox': 'mel_band_roformer_instrumental_bleedless_v3_gabox.ckpt',
        'MelBand Roformer Kim | Inst V1 (E) Plus by Unwa': 'melband_roformer_inst_v1e_plus.ckpt',
        'MelBand Roformer Kim | Inst V1 Plus by Unwa': 'melband_roformer_inst_v1_plus.ckpt',
        'MelBand Roformer Kim | Inst V1 by Unwa': 'melband_roformer_inst_v1.ckpt',
        'MelBand Roformer Kim | Inst V1 (E) by Unwa': 'melband_roformer_inst_v1e.ckpt',
        'MelBand Roformer Kim | Inst V2 by Unwa': 'melband_roformer_inst_v2.ckpt',
        'MelBand Roformer | Instrumental by becruily': 'mel_band_roformer_instrumental_becruily.ckpt',
        'MelBand Roformer | Instrumental by Gabox': 'mel_band_roformer_instrumental_gabox.ckpt',
        'MelBand Roformer | Instrumental 2 by Gabox': 'mel_band_roformer_instrumental_2_gabox.ckpt',
        'MelBand Roformer | Instrumental 3 by Gabox': 'mel_band_roformer_instrumental_3_gabox.ckpt',
        'MelBand Roformer | Instrumental Bleedless V1 by Gabox': 'mel_band_roformer_instrumental_bleedless_v1_gabox.ckpt',
        'MelBand Roformer | Instrumental Bleedless V2 by Gabox': 'mel_band_roformer_instrumental_bleedless_v2_gabox.ckpt',
        'MelBand Roformer | Instrumental Fullness V1 by Gabox': 'mel_band_roformer_instrumental_fullness_v1_gabox.ckpt',
        'MelBand Roformer | Instrumental Fullness V2 by Gabox': 'mel_band_roformer_instrumental_fullness_v2_gabox.ckpt',
        'MelBand Roformer | Instrumental Fullness V3 by Gabox': 'mel_band_roformer_instrumental_fullness_v3_gabox.ckpt',
        'MelBand Roformer | Instrumental Fullness Noisy V4 by Gabox': 'mel_band_roformer_instrumental_fullness_noise_v4_gabox.ckpt',
        'MelBand Roformer | INSTV5 by Gabox': 'mel_band_roformer_instrumental_instv5_gabox.ckpt',
        'MelBand Roformer | INSTV5N by Gabox': 'mel_band_roformer_instrumental_instv5n_gabox.ckpt',
        'MelBand Roformer | INSTV6 by Gabox': 'mel_band_roformer_instrumental_instv6_gabox.ckpt',
        'MelBand Roformer | INSTV6N by Gabox': 'mel_band_roformer_instrumental_instv6n_gabox.ckpt',
        'MelBand Roformer | INSTV7 by Gabox': 'mel_band_roformer_instrumental_instv7_gabox.ckpt',
    },
    "InstVoc Duality": {
        'MelBand Roformer Kim | InstVoc Duality V1 by Unwa': 'melband_roformer_instvoc_duality_v1.ckpt',
        'MelBand Roformer Kim | InstVoc Duality V2 by Unwa': 'melband_roformer_instvox_duality_v2.ckpt',
    },
    "De-Reverb": {
        'BS-Roformer-De-Reverb': 'deverb_bs_roformer_8_384dim_10depth.ckpt',
        'MelBand Roformer | De-Reverb by anvuew': 'dereverb_mel_band_roformer_anvuew_sdr_19.1729.ckpt',
        'MelBand Roformer | De-Reverb Less Aggressive by anvuew': 'dereverb_mel_band_roformer_less_aggressive_anvuew_sdr_18.8050.ckpt',
        'MelBand Roformer | De-Reverb Mono by anvuew': 'dereverb_mel_band_roformer_mono_anvuew.ckpt',
        'MelBand Roformer | De-Reverb Big by Sucial': 'dereverb_big_mbr_ep_362.ckpt',
        'MelBand Roformer | De-Reverb Super Big by Sucial': 'dereverb_super_big_mbr_ep_346.ckpt',
        'MelBand Roformer | De-Reverb-Echo by Sucial': 'dereverb-echo_mel_band_roformer_sdr_10.0169.ckpt',
        'MelBand Roformer | De-Reverb-Echo V2 by Sucial': 'dereverb-echo_mel_band_roformer_sdr_13.4843_v2.ckpt',
        'MelBand Roformer | De-Reverb-Echo Fused by Sucial': 'dereverb_echo_mbr_fused.ckpt',
    },
    "Denoise": {
        'Mel-Roformer-Denoise-Aufr33': 'denoise_mel_band_roformer_aufr33_sdr_27.9959.ckpt',
        'Mel-Roformer-Denoise-Aufr33-Aggr': 'denoise_mel_band_roformer_aufr33_aggr_sdr_27.9768.ckpt',
        'MelBand Roformer | Denoise-Debleed by Gabox': 'mel_band_roformer_denoise_debleed_gabox.ckpt',
        'MelBand Roformer | Bleed Suppressor V1 by unwa-97chris': 'mel_band_roformer_bleed_suppressor_v1.ckpt',
    },
    "Karaoke": {
        'Mel-Roformer-Karaoke-Aufr33-Viperx': 'mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt',
        'MelBand Roformer | Karaoke by Gabox': 'mel_band_roformer_karaoke_gabox.ckpt',
        'MelBand Roformer | Karaoke by becruily': 'mel_band_roformer_karaoke_becruily.ckpt',
    },
    "General Purpose": {
        'BS-Roformer-Viperx-1297': 'model_bs_roformer_ep_317_sdr_12.9755.ckpt',
        'BS-Roformer-Viperx-1296': 'model_bs_roformer_ep_368_sdr_12.9628.ckpt',
        'BS-Roformer-Viperx-1053': 'model_bs_roformer_ep_937_sdr_10.5309.ckpt',
        'Mel-Roformer-Viperx-1143': 'model_mel_band_roformer_ep_3005_sdr_11.4360.ckpt',
        'Mel-Roformer-Crowd-Aufr33-Viperx': 'mel_band_roformer_crowd_aufr33_viperx_sdr_8.7144.ckpt',
        'MelBand Roformer Kim | SYHFT by SYH99999': 'MelBandRoformerSYHFT.ckpt',
        'MelBand Roformer Kim | SYHFT V2 by SYH99999': 'MelBandRoformerSYHFTV2.ckpt',
        'MelBand Roformer Kim | SYHFT V2.5 by SYH99999': 'MelBandRoformerSYHFTV2.5.ckpt',
        'MelBand Roformer Kim | SYHFT V3 by SYH99999': 'MelBandRoformerSYHFTV3Epsilon.ckpt',
        'MelBand Roformer Kim | Big SYHFT V1 by SYH99999': 'MelBandRoformerBigSYHFTV1.ckpt',
        'MelBand Roformer | Aspiration by Sucial': 'aspiration_mel_band_roformer_sdr_18.9845.ckpt',
        'MelBand Roformer | Aspiration Less Aggressive by Sucial': 'aspiration_mel_band_roformer_less_aggr_sdr_18.1201.ckpt',
    }
}

OUTPUT_FORMATS = ['wav', 'flac', 'mp3', 'ogg', 'opus', 'm4a', 'aiff', 'ac3']

# CSS (unchanged)
CSS = """
body {
    background: linear-gradient(to bottom, rgba(45, 11, 11, 0.9), rgba(0, 0, 0, 0.8)), url('/content/logo.jpg') no-repeat center center fixed;
    background-size: cover;
    min-height: 100vh;
    margin: 0;
    padding: 1rem;
    font-family: 'Poppins', sans-serif;
    color: #C0C0C0;
    overflow-x: hidden;
}
.header-text {
    text-align: center;
    padding: 100px 20px 20px;
    color: #ff4040;
    font-size: 3rem;
    font-weight: 900;
    text-shadow: 0 0 10px rgba(255, 64, 64, 0.5);
    z-index: 1500;
    animation: text-glow 2s infinite;
}
.header-subtitle {
    text-align: center;
    color: #C0C0C0;
    font-size: 1.2rem;
    font-weight: 300;
    margin-top: -10px;
    text-shadow: 0 0 5px rgba(255, 64, 64, 0.3);
}
.gr-tab {
    background: rgba(128, 0, 0, 0.5) !important;
    border-radius: 12px 12px 0 0 !important;
    margin: 0 5px !important;
    color: #C0C0C0 !important;
    border: 1px solid #ff4040 !important;
    z-index: 1500;
    transition: background 0.3s ease, color 0.3s ease;
    padding: 10px 20px !important;
    font-size: 1.1rem !important;
}
button {
    transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1) !important;
    background: #800000 !important;
    border: 1px solid #ff4040 !important;
    color: #C0C0C0 !important;
    border-radius: 8px !important;
    padding: 8px 16px !important;
    box-shadow: 0 2px 10px rgba(255, 64, 64, 0.3);
}
button:hover {
    transform: scale(1.05) !important;
    box-shadow: 0 10px 40px rgba(255, 64, 64, 0.7) !important;
    background: #ff4040 !important;
}
.compact-upload.horizontal {
    display: inline-flex !important;
    align-items: center !important;
    gap: 8px !important;
    max-width: 400px !important;
    height: 40px !important;
    padding: 0 12px !important;
    border: 1px solid #ff4040 !important;
    background: rgba(128, 0, 0, 0.5) !important;
    border-radius: 8px !important;
}
.compact-dropdown {
    padding: 8px 12px !important;
    border-radius: 8px !important;
    border: 2px solid #ff6b6b !important;
    background: rgba(46, 26, 71, 0.7) !important;
    color: #e0e0e0 !important;
    width: 100%;
    font-size: 1rem !important;
    transition: border-color 0.3s ease, box-shadow 0.3s ease !important;
    position: relative;
    z-index: 100;
}
.compact-dropdown:hover {
    border-color: #ff8787 !important;
    box-shadow: 0 2px 8px rgba(255, 107, 107, 0.4) !important;
}
.compact-dropdown select, .compact-dropdown .gr-dropdown {
    background: transparent !important;
    color: #e0e0e0 !important;
    border: none !important;
    width: 100% !important;
    padding: 8px !important;
    font-size: 1rem !important;
    appearance: none !important;
    -webkit-appearance: none !important;
    -moz-appearance: none !important;
}
.compact-dropdown .gr-dropdown-menu {
    background: rgba(46, 26, 71, 0.95) !important;
    border: 2px solid #ff6b6b !important;
    border-radius: 8px !important;
    color: #e0e0e0 !important;
    max-height: 300px !important;
    overflow-y: auto !important;
    z-index: 300 !important;
    width: 100% !important;
    opacity: 1 !important;
    visibility: visible !important;
    position: absolute !important;
    top: 100% !important;
    left: 0 !important;
    pointer-events: auto !important;
}
.compact-dropdown:hover .gr-dropdown-menu {
    display: block !important;
}
.compact-dropdown .gr-dropdown-menu option {
    padding: 8px !important;
    color: #e0e0e0 !important;
    background: transparent !important;
}
.compact-dropdown .gr-dropdown-menu option:hover {
    background: rgba(255, 107, 107, 0.3) !important;
}
#custom-progress {
    margin-top: 10px;
    padding: 10px;
    background: rgba(128, 0, 0, 0.3);
    border-radius: 8px;
    border: 1px solid #ff4040;
}
#progress-bar {
    height: 20px;
    background: linear-gradient(to right, #6e8efb, #ff4040);
    border-radius: 5px;
    transition: width 0.5s ease-in-out;
    max-width: 100% !important;
}
.gr-accordion {
    background: rgba(128, 0, 0, 0.5) !important;
    border-radius: 10px !important;
    border: 1px solid #ff4040 !important;
}
.footer {
    text-align: center;
    padding: 20px;
    color: #ff4040;
    font-size: 14px;
    margin-top: 40px;
    background: rgba(128, 0, 0, 0.3);
    border-top: 1px solid #ff4040;
}
#log-accordion {
    max-height: 400px;
    overflow-y: auto;
    background: rgba(0, 0, 0, 0.7) !important;
    padding: 10px;
    border-radius: 8px;
}
@keyframes text-glow {
    0% { text-shadow: 0 0 5px rgba(192, 192, 192, 0); }
    50% { text-shadow: 0 0 15px rgba(192, 192, 192, 1); }
    100% { text-shadow: 0 0 5px rgba(192, 192, 192, 0); }
}
"""

def download_audio(url, cookie_file=None):
    ydl_opts = {
        'format': 'bestaudio[ext=webm]/bestaudio[ext=m4a]/bestaudio[ext=opus]/bestaudio[ext=aac]/bestaudio -video',
        'postprocessors': [{
            'key': 'FFmpegExtractAudio',
            'preferredcodec': 'wav',
            'preferredquality': '192',
        }],
        'outtmpl': 'ytdl/%(title)s.%(ext)s',
        'user_agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
        'geo_bypass': True,
        'force_ipv4': True,
        'referer': 'https://www.youtube.com/',
        'noplaylist': True,
        'cookiefile': cookie_file.name if cookie_file else None,
        'extractor_retries': 5,
        'ignoreerrors': False,
        'no_check_certificate': True,
        'verbose': True,
    }
    temp_output_path = None
    try:
        if 'drive.google.com' in url:
            os.makedirs('ytdl', exist_ok=True)
            file_id = url.split('/d/')[1].split('/')[0]
            download_url = f'https://drive.google.com/uc?id={file_id}'
            temp_output_path = 'ytdl/gdrive_temp_audio'
            gdown.download(download_url, temp_output_path, quiet=False)
            if not os.path.exists(temp_output_path):
                return None, "Downloaded file not found", None
            output_path = 'ytdl/gdrive_audio.wav'
            try:
                audio = AudioSegment.from_file(temp_output_path)
                audio.export(output_path, format="wav")
            except Exception as e:
                return None, f"Failed to process Google Drive file as audio: {str(e)}. Ensure the file contains audio (e.g., MP3, WAV, or video with audio track).", None
            sample_rate, data = scipy.io.wavfile.read(output_path)
            return output_path, "Download and audio conversion successful", (sample_rate, data)
        else:
            os.makedirs('ytdl', exist_ok=True)
            with yt_dlp.YoutubeDL(ydl_opts) as ydl:
                info_dict = ydl.extract_info(url, download=True)
                base_file_path = ydl.prepare_filename(info_dict)
                file_path = base_file_path
                for ext in ['.webm', '.m4a', '.opus', '.aac']:
                    file_path = file_path.replace(ext, '.wav')
                if not os.path.exists(file_path):
                    return None, "Downloaded file not found", None
                sample_rate, data = scipy.io.wavfile.read(file_path)
                return file_path, "Download successful", (sample_rate, data)
    except yt_dlp.utils.ExtractorError as e:
        if "Sign in to confirm you’re not a bot" in str(e):
            return None, "Authentication error. Please upload valid YouTube cookies: https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies", None
        return None, f"Download error: {str(e)}", None
    except Exception as e:
        return None, f"Unexpected error: {str(e)}", None
    finally:
        if temp_output_path and os.path.exists(temp_output_path):
            os.remove(temp_output_path)
            logger.info(f"Temporary file deleted: {temp_output_path}")

@spaces.GPU
def roformer_separator(audio, model_key, seg_size, override_seg_size, overlap, pitch_shift, model_dir, output_dir, out_format, norm_thresh, amp_thresh, batch_size, exclude_stems="", progress=gr.Progress(track_tqdm=True)):
    if not audio:
        raise ValueError("No audio file provided.")
    temp_audio_path = None
    try:
        if isinstance(audio, tuple):
            sample_rate, data = audio
            temp_audio_path = os.path.join("/tmp", "temp_audio.wav")
            scipy.io.wavfile.write(temp_audio_path, sample_rate, data)
            audio = temp_audio_path
        if seg_size > 512:
            logger.warning(f"Segment size {seg_size} is large, this may cause issues.")
        override_seg_size = override_seg_size == "True"
        if os.path.exists(output_dir):
            shutil.rmtree(output_dir)
        os.makedirs(output_dir, exist_ok=True)
        base_name = os.path.splitext(os.path.basename(audio))[0]
        for category, models in ROFORMER_MODELS.items():
            if model_key in models:
                model = models[model_key]
                break
        else:
            raise ValueError(f"Model '{model_key}' not found.")
        logger.info(f"Separating {base_name} with {model_key} on {device}")
        separator = Separator(
            log_level=logging.INFO,
            model_file_dir=model_dir,
            output_dir=output_dir,
            output_format=out_format,
            normalization_threshold=norm_thresh,
            amplification_threshold=amp_thresh,
            use_autocast=use_autocast,
            mdxc_params={"segment_size": seg_size, "override_model_segment_size": override_seg_size, "batch_size": batch_size, "overlap": overlap, "pitch_shift": pitch_shift}
        )
        progress(0.2, desc="Loading model...")
        separator.load_model(model_filename=model)
        progress(0.7, desc="Separating audio...")
        separation = separator.separate(audio)
        stems = [os.path.join(output_dir, file_name) for file_name in separation]
        file_list = []
        if exclude_stems.strip():
            excluded = [s.strip().lower() for s in exclude_stems.split(',')]
            filtered_stems = [stem for stem in stems if not any(ex in os.path.basename(stem).lower() for ex in excluded)]
            file_list = filtered_stems
            stem1 = filtered_stems[0] if filtered_stems else None
            stem2 = filtered_stems[1] if len(filtered_stems) > 1 else None
        else:
            file_list = stems
            stem1 = stems[0]
            stem2 = stems[1] if len(stems) > 1 else None
        return stem1, stem2, file_list
    except Exception as e:
        logger.error(f"Separation error: {e}")
        raise RuntimeError(f"Separation error: {e}")
    finally:
        if temp_audio_path and os.path.exists(temp_audio_path):
            os.remove(temp_audio_path)
            logger.info(f"Temporary file deleted: {temp_audio_path}")
        if torch.cuda.is_available():
            torch.cuda.empty_cache()
            logger.info("GPU memory cleared")

@spaces.GPU
def auto_ensemble_process(audio, model_keys, seg_size=64, overlap=0.1, out_format="wav", use_tta="False", model_dir="/tmp/audio-separator-models/", output_dir="output", norm_thresh=0.9, amp_thresh=0.9, batch_size=1, ensemble_method="avg_wave", exclude_stems="", weights_str="", progress=gr.Progress(track_tqdm=True)):
    temp_audio_path = None
    max_retries = 2
    start_time = time.time()
    time_budget = 100  # seconds
    max_models = 6
    gpu_lock = Lock()

    try:
        if not audio:
            raise ValueError("No audio file provided.")
        if not model_keys:
            raise ValueError("No models selected.")
        if len(model_keys) > max_models:
            logger.warning(f"Selected {len(model_keys)} models, limiting to {max_models}.")
            model_keys = model_keys[:max_models]

        # Dynamic batch size based on audio duration and model count
        audio_data, sr = librosa.load(audio, sr=None, mono=False)
        duration = librosa.get_duration(y=audio_data, sr=sr)
        logger.info(f"Audio duration: {duration:.2f} seconds")
        dynamic_batch_size = max(1, min(4, 1 + int(900 / (duration + 1)) - len(model_keys) // 2))
        logger.info(f"Using batch size: {dynamic_batch_size} for {len(model_keys)} models, duration {duration:.2f}s")

        if isinstance(audio, tuple):
            sample_rate, data = audio
            temp_audio_path = os.path.join("/tmp", "temp_audio.wav")
            scipy.io.wavfile.write(temp_audio_path, sample_rate, data)
            audio = temp_audio_path

        use_tta = use_tta == "True"
        if os.path.exists(output_dir):
            shutil.rmtree(output_dir)
        os.makedirs(output_dir, exist_ok=True)
        base_name = os.path.splitext(os.path.basename(audio))[0]
        logger.info(f"Ensemble for {base_name} with {model_keys} on {device}")

        # Model cache
        model_cache = {}
        all_stems = []
        model_stems = {model_key: {"vocals": [], "other": []} for model_key in model_keys}
        total_tasks = len(model_keys)

        def process_model(model_key, model_idx):
            with torch.no_grad():
                for attempt in range(max_retries + 1):
                    try:
                        # Find model
                        for category, models in ROFORMER_MODELS.items():
                            if model_key in models:
                                model = models[model_key]
                                break
                        else:
                            logger.warning(f"Model {model_key} not found, skipping")
                            return []

                        # Check time budget
                        elapsed = time.time() - start_time
                        if elapsed > time_budget:
                            logger.error(f"Time budget ({time_budget}s) exceeded")
                            raise TimeoutError("Processing took too long")

                        # Initialize separator
                        model_path = os.path.join(model_dir, model)
                        if model_key not in model_cache:
                            logger.info(f"Loading {model_key} into cache")
                            separator = Separator(
                                log_level=logging.INFO,
                                model_file_dir=model_dir,
                                output_dir=output_dir,
                                output_format=out_format,
                                normalization_threshold=norm_thresh,
                                amplification_threshold=amp_thresh,
                                use_autocast=use_autocast,
                                mdxc_params={
                                    "segment_size": seg_size,
                                    "overlap": overlap,
                                    "use_tta": use_tta,
                                    "batch_size": dynamic_batch_size
                                }
                            )
                            separator.load_model(model_filename=model)
                            model_cache[model_key] = separator
                        else:
                            separator = model_cache[model_key]

                        # Process with GPU lock
                        with gpu_lock:
                            progress(0.3 + (model_idx / total_tasks) * 0.5, desc=f"Separating with {model_key}")
                            logger.info(f"Separating with {model_key}")
                            separation = separator.separate(audio)
                            stems = [os.path.join(output_dir, file_name) for file_name in separation]
                            result = []
                            for stem in stems:
                                if "vocals" in os.path.basename(stem).lower():
                                    model_stems[model_key]["vocals"].append(stem)
                                elif "other" in os.path.basename(stem).lower() or "instrumental" in os.path.basename(stem).lower():
                                    model_stems[model_key]["other"].append(stem)
                                    result.append(stem)
                            return result
                    except Exception as e:
                        logger.error(f"Error processing {model_key}, attempt {attempt + 1}/{max_retries + 1}: {e}")
                        if attempt == max_retries:
                            logger.error(f"Max retries reached for {model_key}, skipping")
                            return []
                        time.sleep(1)
                    finally:
                        if torch.cuda.is_available():
                            torch.cuda.empty_cache()
                            logger.info(f"Cleared CUDA cache after {model_key}")

        # Parallel processing
        progress(0.1, desc="Starting model separations...")
        with ThreadPoolExecutor(max_workers=min(4, len(model_keys))) as executor:
            future_to_task = {executor.submit(process_model, model_key, idx): model_key for idx, model_key in enumerate(model_keys)}
            for future in as_completed(future_to_task):
                model_key = future_to_task[future]
                try:
                    stems = future.result()
                    if stems:
                        logger.info(f"Completed {model_key}")
                    else:
                        logger.warning(f"No stems produced for {model_key}")
                except Exception as e:
                    logger.error(f"Task {model_key} failed: {e}")

        # Clear model cache
        model_cache.clear()
        gc.collect()
        if torch.cuda.is_available():
            torch.cuda.empty_cache()
            logger.info("Cleared model cache and GPU memory")

        # Combine stems
        progress(0.8, desc="Combining stems...")
        for model_key, stems_dict in model_stems.items():
            for stem_type in ["vocals", "other"]:
                if stems_dict[stem_type]:
                    combined_path = os.path.join(output_dir, f"{base_name}_{stem_type}_{model_key.replace(' | ', '_').replace(' ', '_')}.wav")
                    try:
                        data, _ = librosa.load(stems_dict[stem_type][0], sr=sr, mono=False)
                        with sf.SoundFile(combined_path, 'w', sr, channels=2 if data.ndim == 2 else 1) as f:
                            f.write(data.T if data.ndim == 2 else data)
                        logger.info(f"Combined {stem_type} for {model_key}: {combined_path}")
                        if exclude_stems.strip() and stem_type.lower() in [s.strip().lower() for s in exclude_stems.split(',')]:
                            logger.info(f"Excluding {stem_type} for {model_key}")
                            continue
                        all_stems.append(combined_path)
                    except Exception as e:
                        logger.error(f"Error combining {stem_type} for {model_key}: {e}")

        all_stems = [stem for stem in all_stems if os.path.exists(stem)]
        if not all_stems:
            raise ValueError("No valid stems found for ensemble. Try uploading a local WAV file.")

        # Ensemble
        weights = [float(w.strip()) for w in weights_str.split(',')] if weights_str.strip() else [1.0] * len(all_stems)
        if len(weights) != len(all_stems):
            weights = [1.0] * len(all_stems)
            logger.info("Weights mismatched, defaulting to 1.0")
        output_file = os.path.join(output_dir, f"{base_name}_ensemble_{ensemble_method}.{out_format}")
        ensemble_args = [
            "--files", *all_stems,
            "--type", ensemble_method,
            "--weights", *[str(w) for w in weights],
            "--output", output_file
        ]
        progress(0.9, desc="Running ensemble...")
        logger.info(f"Running ensemble with args: {ensemble_args}")
        try:
            result = ensemble_files(ensemble_args)
            if result is None or not os.path.exists(output_file):
                raise RuntimeError(f"Ensemble failed, output file not created: {output_file}")
            logger.info(f"Ensemble completed, output: {output_file}")
            progress(1.0, desc="Ensemble completed")
            elapsed = time.time() - start_time
            logger.info(f"Total processing time: {elapsed:.2f}s")
            # Prepare file list for download
            file_list = [output_file] + all_stems
            # Create status message with download links
            status = f"Ensemble completed with {ensemble_method}, excluded: {exclude_stems if exclude_stems else 'None'}, {len(model_keys)} models in {elapsed:.2f}s<br>Download files:<ul>"
            for file in file_list:
                file_name = os.path.basename(file)
                status += f"<li><a href='file={file}' download>{file_name}</a></li>"
            status += "</ul>"
            return output_file, status, file_list
        except Exception as e:
            logger.error(f"Ensemble processing error: {e}")
            if "numpy" in str(e).lower() or "copy" in str(e).lower():
                error_msg = f"NumPy compatibility error: {e}. Try installing numpy<2.0.0 or contact support."
            else:
                error_msg = f"Ensemble processing error: {e}"
            raise RuntimeError(error_msg)
    except Exception as e:
        logger.error(f"Ensemble error: {e}")
        error_msg = f"Processing failed. Try fewer models (max {max_models}), shorter audio, or uploading a local WAV file."
        raise RuntimeError(error_msg)
    finally:
        if temp_audio_path and os.path.exists(temp_audio_path):
            try:
                os.remove(temp_audio_path)
                logger.info(f"Temporary file deleted: {temp_audio_path}")
            except Exception as e:
                logger.warning(f"Failed to delete temporary file {temp_audio_path}: {e}")
        if torch.cuda.is_available():
            torch.cuda.empty_cache()
            logger.info("GPU memory cleared")

def update_roformer_models(category):
    """Update Roformer model dropdown based on selected category."""
    choices = list(ROFORMER_MODELS.get(category, {}).keys()) or []
    logger.debug(f"Updating roformer models for category {category}: {choices}")
    return gr.update(choices=choices, value=choices[0] if choices else None)

def update_ensemble_models(category):
    """Update ensemble model dropdown based on selected category."""
    choices = list(ROFORMER_MODELS.get(category, {}).keys()) or []
    logger.debug(f"Updating ensemble models for category {category}: {choices}")
    return gr.update(choices=choices, value=[])

def download_audio_wrapper(url, cookie_file):
    file_path, status, audio_data = download_audio(url, cookie_file)
    return audio_data, status

def create_interface():
    with gr.Blocks(title="🎵 SESA Fast Separation 🎵", css=CSS, elem_id="app-container") as app:
        gr.Markdown("<h1 class='header-text'>🎵 SESA Fast Separation 🎵</h1>")
        gr.Markdown("**Note**: If YouTube downloads fail, upload a valid cookies file or a local WAV file. [Cookie Instructions](https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies)")
        gr.Markdown("**Tip**: For best results, use audio shorter than 15 minutes or fewer models (up to 6) to ensure smooth processing.")
        with gr.Tabs():
            with gr.Tab("⚙️ Settings"):
                with gr.Group(elem_classes="dubbing-theme"):
                    gr.Markdown("### General Settings")
                    model_file_dir = gr.Textbox(value="/tmp/audio-separator-models/", label="📂 Model Cache", placeholder="Path to model directory", interactive=True)
                    output_dir = gr.Textbox(value="output", label="📤 Output Directory", placeholder="Where to save results", interactive=True)
                    output_format = gr.Dropdown(value="wav", choices=OUTPUT_FORMATS, label="🎶 Output Format", interactive=True)
                    norm_threshold = gr.Slider(0.1, 1.0, value=0.9, step=0.1, label="🔊 Normalization Threshold", interactive=True)
                    amp_threshold = gr.Slider(0.1, 1.0, value=0.3, step=0.1, label="📈 Amplification Threshold", interactive=True)
                    batch_size = gr.Slider(1, 8, value=1, step=1, label="⚡ Batch Size", interactive=True)
            with gr.Tab("🎤 Roformer"):
                with gr.Group(elem_classes="dubbing-theme"):
                    gr.Markdown("### Audio Separation")
                    with gr.Row():
                        roformer_audio = gr.Audio(label="🎧 Upload Audio", type="filepath", interactive=True)
                        url_ro = gr.Textbox(label="🔗 Or Paste URL", placeholder="YouTube or audio URL", interactive=True)
                        cookies_ro = gr.File(label="🍪 Cookies File", file_types=[".txt"], interactive=True)
                        download_roformer = gr.Button("⬇️ Download", variant="secondary")
                    roformer_download_status = gr.Textbox(label="📢 Download Status", interactive=False)
                    roformer_exclude_stems = gr.Textbox(label="🚫 Exclude Stems", placeholder="e.g., vocals, drums (comma-separated)", interactive=True)
                    with gr.Row():
                        roformer_category = gr.Dropdown(label="📚 Category", choices=list(ROFORMER_MODELS.keys()), value="General Purpose", interactive=True)
                        roformer_model = gr.Dropdown(label="🛠️ Model", choices=list(ROFORMER_MODELS["General Purpose"].keys()), interactive=True, allow_custom_value=True)
                    with gr.Row():
                        roformer_seg_size = gr.Slider(32, 512, value=64, step=32, label="📏 Segment Size", interactive=True)
                        roformer_overlap = gr.Slider(2, 10, value=8, step=1, label="🔄 Overlap", interactive=True)
                    with gr.Row():
                        roformer_pitch_shift = gr.Slider(-12, 12, value=0, step=1, label="🎵 Pitch Shift", interactive=True)
                        roformer_override_seg_size = gr.Dropdown(choices=["True", "False"], value="False", label="🔧 Override Segment Size", interactive=True)
                    roformer_button = gr.Button("✂️ Separate Now!", variant="primary")
                    with gr.Row():
                        roformer_stem1 = gr.Audio(label="🎸 Stem 1", type="filepath", interactive=False)
                        roformer_stem2 = gr.Audio(label="🥁 Stem 2", type="filepath", interactive=False)
                    roformer_files = gr.File(label="📥 Download Stems", interactive=False)
            with gr.Tab("🎚️ Auto Ensemble"):
                with gr.Group(elem_classes="dubbing-theme"):
                    gr.Markdown("### Ensemble Processing")
                    gr.Markdown("Note: If weights are not specified, equal weights (1.0) are applied. Use up to 6 models for best results.")
                    with gr.Row():
                        ensemble_audio = gr.Audio(label="🎧 Upload Audio", type="filepath", interactive=True)
                        url_ensemble = gr.Textbox(label="🔗 Or Paste URL", placeholder="YouTube or audio URL", interactive=True)
                        cookies_ensemble = gr.File(label="🍪 Cookies File", file_types=[".txt"], interactive=True)
                        download_ensemble = gr.Button("⬇️ Download", variant="secondary")
                    ensemble_download_status = gr.Textbox(label="📢 Download Status", interactive=False)
                    ensemble_exclude_stems = gr.Textbox(label="🚫 Exclude Stems", placeholder="e.g., vocals, drums (comma-separated)", interactive=True)
                    with gr.Row():
                        ensemble_category = gr.Dropdown(label="📚 Category", choices=list(ROFORMER_MODELS.keys()), value="Instrumentals", interactive=True)
                        ensemble_models = gr.Dropdown(label="🛠️ Models (Max 6)", choices=list(ROFORMER_MODELS["Instrumentals"].keys()), multiselect=True, interactive=True, allow_custom_value=True)
                    with gr.Row():
                        ensemble_seg_size = gr.Slider(32, 512, value=64, step=32, label="📏 Segment Size", interactive=True)
                        ensemble_overlap = gr.Slider(2, 10, value=8, step=1, label="🔄 Overlap", interactive=True)
                        ensemble_use_tta = gr.Dropdown(choices=["True", "False"], value="False", label="🔍 Use TTA", interactive=True)
                    ensemble_method = gr.Dropdown(label="⚙️ Ensemble Method", choices=['avg_wave', 'median_wave', 'max_wave', 'min_wave', 'avg_fft', 'median_fft', 'max_fft', 'min_fft'], value='avg_wave', interactive=True)
                    ensemble_weights = gr.Textbox(label="⚖️ Weights", placeholder="e.g., 1.0, 1.0, 1.0 (comma-separated)", interactive=True)
                    ensemble_button = gr.Button("🎛️ Run Ensemble!", variant="primary")
                    ensemble_output = gr.Audio(label="🎶 Ensemble Result", type="filepath", interactive=False)
                    ensemble_status = gr.HTML(label="📢 Status")
                    ensemble_files = gr.File(label="📥 Download Ensemble and Stems", interactive=False)
        gr.HTML("<div class='footer'>Powered by Audio-Separator 🌟🎶 | Made with ❤️</div>")
        roformer_category.change(update_roformer_models, inputs=[roformer_category], outputs=[roformer_model])
        download_roformer.click(
            fn=download_audio_wrapper,
            inputs=[url_ro, cookies_ro],
            outputs=[roformer_audio, roformer_download_status]
        )
        roformer_button.click(
            fn=roformer_separator,
            inputs=[
                roformer_audio, roformer_model, roformer_seg_size, roformer_override_seg_size,
                roformer_overlap, roformer_pitch_shift, model_file_dir, output_dir,
                output_format, norm_threshold, amp_threshold, batch_size, roformer_exclude_stems
            ],
            outputs=[roformer_stem1, roformer_stem2, roformer_files]
        )
        ensemble_category.change(update_ensemble_models, inputs=[ensemble_category], outputs=[ensemble_models])
        download_ensemble.click(
            fn=download_audio_wrapper,
            inputs=[url_ensemble, cookies_ensemble],
            outputs=[ensemble_audio, ensemble_download_status]
        )
        ensemble_button.click(
            fn=auto_ensemble_process,
            inputs=[
                ensemble_audio, ensemble_models, ensemble_seg_size, ensemble_overlap,
                output_format, ensemble_use_tta, model_file_dir, output_dir,
                norm_threshold, amp_threshold, batch_size, ensemble_method,
                ensemble_exclude_stems, ensemble_weights
            ],
            outputs=[ensemble_output, ensemble_status, ensemble_files]
        )
    return app

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Music Source Separation Web UI")
    parser.add_argument("--port", type=int, default=7860, help="Port to run the UI on")
    args = parser.parse_args()
    app = create_interface()
    try:
        app.launch(server_name="0.0.0.0", server_port=args.port, share=True)
    except Exception as e:
        logger.error(f"Failed to launch UI: {e}")
        raise
    finally:
        app.close()