zlei6/OpenMontage
1
0
Fork 0
mirror of https://github.com/calesthio/OpenMontage.git synced 2026-08-24 17:10:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(audio_mixer): asplit speech stream so ducking graph is CI-portable

Browse Source 
The prior fix removed the dangling pad but still reused the speech filter
output for two consumers (sidechain key + final mix). FFmpeg auto-splits a
reused *input* label on some builds (macOS) but the Linux ffmpeg on CI rejects
it, so both full_mix ducking tests failed there.

Build a single [speech_all] stream and asplit it into [speech_key] (sidechain
key) and [speech_out] (final mix) so every filter label is produced once and
consumed once. Verified the generated graph for the single- and multi-narration
cases: no label is consumed more than once.

Refs #265
This commit is contained in:
0xDevNinja
2026-07-03 12:58:29 +05:30
parent f08a77979c
commit 6426662083
1 changed files with 12 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
tools/audio/audio_mixer.py
View File

@@ -492,17 +492,22 @@ class AudioMixer(BaseTool):
duck_enabled = ducking.get("enabled", True) if isinstance(ducking, dict) else bool(ducking)
if duck_enabled and speech_tracks and music_tracks:
# Mix speech tracks together first
# Build ONE speech stream, then split it into two independent
# branches: one feeds the sidechain compressor as the ducking key,
# the other is mixed into the final output. A filtergraph label may
# only be consumed once, so reusing the same speech label for both
# the sidechain key and the output mix is invalid on stricter ffmpeg
# builds (e.g. the Linux ffmpeg on CI). asplit makes the fork explicit.
speech_indices = list(range(len(speech_tracks)))
speech_labels = "".join(f"[a{i}]" for i in speech_indices)
if len(speech_tracks) > 1:
filter_parts.append(
f"{speech_labels}amix=inputs={len(speech_tracks)}:duration=longest[speech_mix]"
f"{speech_labels}amix=inputs={len(speech_tracks)}:duration=longest[speech_all]"
)
speech_out = "[speech_mix]"
else:
speech_out = f"[a{speech_indices[0]}]"
filter_parts.append(f"[a{speech_indices[0]}]acopy[speech_all]")
filter_parts.append("[speech_all]asplit=2[speech_key][speech_out]")
# Mix music tracks together
music_start = len(speech_tracks)
@@ -517,35 +522,20 @@ class AudioMixer(BaseTool):
else:
music_in = f"[a{music_indices[0]}]"
# Apply sidechain ducking
# Apply sidechain ducking — music is compressed, [speech_key] is the key
duck_params = ducking if isinstance(ducking, dict) else {}
attack = duck_params.get("attack_ms", 200) / 1000
release = duck_params.get("release_ms", 500) / 1000
music_vol = duck_params.get("music_volume_during_speech", 0.15)
filter_parts.append(
f"{music_in}{speech_out}sidechaincompress="
f"{music_in}[speech_key]sidechaincompress="
f"threshold=0.02:ratio=9:attack={attack}:release={release}:"
f"level_sc=1:mix=0.9[ducked_music];"
f"[ducked_music]volume={music_vol * 3}[music_out]"
)
# sidechaincompress uses the speech signal only as the ducking key —
# it does not emit speech to the output. Re-derive the speech stream
# for the final mix below. (An earlier version also appended an
# `acopy[speech_dup]` here, but that pad was never consumed and left
# the filtergraph with a dangling output, which ffmpeg rejects — so
# single-narration + music full_mix always failed. FFmpeg auto-splits
# the reused input label, so no explicit duplicate is needed.)
# Build speech mix for output separately
if len(speech_tracks) > 1:
# speech_mix already exists, make a copy for output
filter_parts.append(f"{speech_labels}amix=inputs={len(speech_tracks)}:duration=longest[speech_out]")
else:
filter_parts.append(f"[a{speech_indices[0]}]acopy[speech_out]")
# Final mix: speech_out + music_out
# Final mix: the other speech branch + ducked music
mix_label = "[speech_out][music_out]amix=inputs=2:duration=longest[premix]"
# Add SFX if present