whisper:tldr:aca13

whisper: Convert an audio file using a specific model for conversion.

whisper

$ whisper ${path-to-audio-mp3} --model ${tiny-en,tiny,base-en,base,small-en,small,medium-en,medium,large-v1,large-v2,large}

try on your machine

This command is using the "whisper" command-line tool to perform speech recognition on an audio file (in mp3 format) specified by the variable "${path-to-audio-mp3}". The tool utilizes different models for speech recognition, and the specific model to be used is determined by the value of the variable "${tiny-en,tiny,base-en,base,small-en,small,medium-en,medium,large-v1,large-v2,large}".

The models vary in terms of size and complexity, as well as their trained capabilities to recognize different languages and accents. Each option in the "${tiny-en,tiny,base-en,base,small-en,small,medium-en,medium,large-v1,large-v2,large}" variable corresponds to a different model. The available options include tiny, base, small, medium, and large models, with some being fine-tuned for English specifically ("en" appended to the name).

By replacing "${path-to-audio-mp3}" with the actual path to an audio mp3 file and "${tiny-en,tiny,base-en,base,small-en,small,medium-en,medium,large-v1,large-v2,large}" with the desired model option, the command will recognize speech in the given audio file using the specified model.

This explanation was created by an AI. In most cases those are correct. But please always be careful and never run a command you are not sure if it is safe.

back to the whisper tool