Using audio threshold and custom silent speed together makes rendering super slow

[ErikVogelLUH]

Bug description

When you define a custom audio threshold and a custom silent speed, exporting the video becomes super slow. I always use h264_nvenc together with --my-ffmpeg, because else it is even slower. The ETA goes way up during the encoding process. In the task manager I can see that my GPU only has small spikes of video encoding. With default option auto-editor fully utilizes my GPU. Over the process of the rendering I can see that the spikes become less frequent and more spaced out. At the same time the ETA goes up

What's your platform?

Python Version: 3.11.4 64-bit
Platform: Windows 10 amd64
FFmpeg Version: N-110322-g0c25da5a28-g1440bf15e2+1
FFmpeg Path: C:\Program Files\Python311\Lib\site-packages\ae_ffmpeg\Windows\ffmpeg.exe
Auto-Editor Version: 24w19a

What command did you use

auto-editor '.\video.webm' -c:v h264_nvenc --no-open --my-ffmpeg -s 5 --edit "audio:threshold=7%" -o video_out.mkv

What properties does your input video have?

• video:
  • track 0:
    • codec: vp9
    • fps: 24000/1001
    • resolution: 1920x1080
    • aspect ratio: 16:9
    • pixel aspect ratio: 1
    • pix fmt: yuv420p
    • color range: 1 (tv)
    • color space: 1 (bt709)
    • color primaries: 1 (bt709)
    • color transfer: 1 (bt709)
    • timebase: 1/1000
    • lang: eng
• audio:
  • track 0:
    • codec: opus
    • samplerate: 48000
    • channels: 2
    • lang: eng
• container:
  • duration: 1545.988
  • bitrate: 261192

  • Input #0, matroska,webm, from '.\video.webm': Metadata: ENCODER : Lavf60.6.100 Duration: 00:25:45.99, start: 0.000000, bitrate: 261 kb/s Stream #0:0(eng): Video: vp9 (Profile 0), yuv420p(tv, bt709), 1920x1080, SAR 1:1 DAR 16:9, 23.98 fps, 23.98 tbr, 1k tbn (default) Metadata: DURATION : 00:25:45.961000000 Stream #0:1(eng): Audio: opus, 48000 Hz, stereo, fltp (default) Metadata: DURATION : 00:25:45.988000000

Comments

No response

[WyattBlue]

I can replicate this, will figure out a fix soon.

[WyattBlue]

2612980838ca3495b0845274177a3ff57e679e5c fixes this.