Files
2026-03-03 14:52:19 +09:00

97 lines
3.0 KiB
Python
Executable File
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""
Preset mixing: combine separated stems into a single output file.
- Instrumental: all non-vocal stems mixed
- Static: all stems except user-selected ones mixed
- Dynamic: all stems mixed with per-stem volume multipliers (0.0–2.0)
"""
import math
from pathlib import Path
from typing import Dict, List
from pydub import AudioSegment
def _load(path: str) -> AudioSegment:
return AudioSegment.from_file(path)
def _apply_gain(audio: AudioSegment, multiplier: float) -> AudioSegment:
"""Convert a linear multiplier to dB and apply as gain."""
if multiplier <= 0.0:
return AudioSegment.silent(duration=len(audio), frame_rate=audio.frame_rate)
gain_db = 20 * math.log10(multiplier)
return audio + gain_db
def _export(audio: AudioSegment, dest: Path, output_format: str) -> None:
if output_format == 'mp3_320':
audio.export(str(dest), format='mp3', bitrate='320k')
elif output_format == 'mp3':
audio.export(str(dest), format='mp3', bitrate='256k')
elif output_format == 'flac':
audio.export(str(dest), format='flac')
else:
audio.export(str(dest), format='wav')
def _mix(paths: List[str]) -> AudioSegment:
"""Overlay all audio files into one mix."""
if not paths:
raise ValueError("No stems to mix")
result = _load(paths[0])
for p in paths[1:]:
result = result.overlay(_load(p))
return result
def make_instrumental(
stem_files: Dict[str, str],
output_path: Path,
output_format: str,
) -> str:
"""Mix all stems except those with 'vocal' in the stem name."""
keep = [p for name, p in stem_files.items() if 'vocal' not in name.lower()]
if not keep:
raise RuntimeError("No non-vocal stems available for Instrumental preset")
_export(_mix(keep), output_path, output_format)
return str(output_path)
def make_static(
stem_files: Dict[str, str],
muted_stems: List[str],
output_path: Path,
output_format: str,
) -> str:
"""Mix all stems except those listed in muted_stems."""
muted_lower = {s.lower() for s in muted_stems}
keep = [p for name, p in stem_files.items() if name.lower() not in muted_lower]
if not keep:
raise RuntimeError("All stems are muted — nothing to mix for Static preset")
_export(_mix(keep), output_path, output_format)
return str(output_path)
def make_dynamic(
stem_files: Dict[str, str],
volumes: Dict[str, float],
output_path: Path,
output_format: str,
) -> str:
"""
Mix stems with per-stem linear volume multipliers.
volumes keys are stem names (lowercase); missing stems default to 1.0.
"""
adjusted: List[AudioSegment] = []
for name, path in stem_files.items():
vol = volumes.get(name.lower(), 1.0)
adjusted.append(_apply_gain(_load(path), vol))
if not adjusted:
raise RuntimeError("No stems available for Dynamic preset")
result = adjusted[0]
for a in adjusted[1:]:
result = result.overlay(a)
_export(result, output_path, output_format)
return str(output_path)