search

javedali99 / audio-to-text-transcription

This repository contains a Python script that allows users to download the audio from a YouTube video, transcribe it into text, detect the language and save the transcription in txt file automatically.
https://www.javedali.net/post/2023-04-audio-to-text/
MIT License
117 stars 18 forks source link

help please!!! #1

Closed KENPOKASMMA1982 closed 1 year ago

KENPOKASMMA1982 commented 1 year ago

C:\Users\User\Downloads\audio-to-text-transcription-main\audio-to-text-transcription-main>python youtube_audio_to_text.py Enter the YouTube video URL: https://www.youtube.com/watch?v=r9h-0lDrTdA&t=5s Audio downloaded to YoutubeAudios/audio.mp3 C:\Users\User\AppData\Roaming\Python\Python311\site-packages\whisper\transcribe.py:114: UserWarning: FP16 is not supported on CPU; using FP32 instead warnings.warn("FP16 is not supported on CPU; using FP32 instead")

javedali99 commented 1 year ago

@KENPOKASMMA1982 The message you're seeing is a warning, not an error. It originates from the whisper module the script uses for transcription. The warning indicates that the module attempted to use 16-bit Floating Point (FP16) precision, which is often optimized for certain GPUs. However, since it's running on a CPU that doesn't support FP16, it defaults to 32-bit Floating Point (FP32) precision instead. This might result in slightly slower performance, but the script should continue to function without issues.

KENPOKASMMA1982 commented 1 year ago

hello my friend!!!could you tell me step ,by step what I have to do to use your amazing project?

javedali99 commented 1 year ago

@KENPOKASMMA1982 Check this blog post for the instructions: https://www.javedali.net/post/2023-04-audio-to-text/