解决CMD环境下FFmpeg静默检测结果的多行输出写入文件问题

CMD Command (Single Line, Overwrite Mode) #

This approach starts with a clean file and writes all matching results in one overwrite operation:

del "C:\TEMP.txt" 2>nul & ffmpeg -i "C:\any-video.avi" -af silencedetect=noise=-50dB:d=0.1 -f null - 2>&1 | findstr "silence_duration:" | (for /f "tokens=8" %f in ('more') do @echo %f) > "C:\TEMP.txt"

Breakdown: #

• del "C:\TEMP.txt" 2>nul: Deletes the existing file (suppresses errors if it doesn’t exist)
• ffmpeg ... 2>&1: Runs ffmpeg and redirects its stderr output (where silencedetect logs data) to stdout
• findstr "silence_duration:": Filters output to only lines containing silence duration values
• Wrapping the for /f loop in parentheses and using > at the end redirects all echoed lines to the file in one go, preserving every result instead of overwriting per line

PowerShell Command (Simpler Native Overwrite) #

PowerStream’s stream processing handles multi-line output natively, making it easier to capture all results and overwrite the file by default:

ffmpeg -i "C:\any-video.avi" -af silencedetect=noise=-50dB:d=0.1 -f null - 2>&1 | Select-String "silence_duration:" | ForEach-Object { $_.Line.Split()[7] } | Out-File "C:\TEMP.txt" -Encoding Unicode

Breakdown: #

• ffmpeg ... 2>&1: Redirects ffmpeg’s stderr to stdout, same as CMD
• Select-String "silence_duration:": Filters for lines with silence duration data
• ForEach-Object { $_.Line.Split()[7] }: Splits each line into tokens and extracts the 8th value (zero-based index 7)
• Out-File ...: Writes all results to the file, overwriting existing content by default. -Encoding Unicode matches CMD’s /U output format if needed

内容的提问来源于stack exchange，提问作者DannyhelMont