Transcribing Audio with Whisper CLI | Generated by AI
File and whisper CLI are both present. Basic command:
whisper "meeting_01 2.mp4" --model small --output_format txt --output_dir .
Useful tweaks:
--model—tiny,base,small,medium,large-v3(bigger = slower, more accurate)--language en— skip auto-detection if you know the language--output_format—txt,srt,vtt,json, orall--device cuda— if you have a GPU (default is CPU)--fp16 False— required on CPU to avoid a warning
CPU-friendly example:
whisper "meeting_01 2.mp4" --model small --language en --fp16 False --output_format txt --output_dir .
This writes meeting_01 2.txt (and same basename for other formats) into the current directory. Want me to run it?
Generated by AI. Curating and sharing still takes effort.
If you find it useful, feel free to
donate.
WeChat: @lzwjava
·
X: @lzwjava ·
Say hi 👋