Transcribe Audio Files with OpenAI Whisper

In this video, we learn how to transcribe audio files with OpenAI whisper in Python.

◾◾◾◾◾◾◾◾◾◾◾◾◾◾◾◾◾
📚 Programming Books & Merch 📚

🌐 Social Media & Contact 🌐

[nicholaspitti]

For anyone encountering "UserWarning: FP16 is not supported on CPU; using FP32 instead" add "fp16=False" as a parameter in the transcribe function -> model.transcribe("audio.mp3", fp16=False)

[SmokeytheBeer]

FYI for anyone trying this themselves: when installing the open ai whisper package from pip, the package name is "openai-whisper". There is already another packaged called "whisper" that is for a fixed size round-robin style database. However, the import statement used in this video is correct.

[the_huge_knight]

I recommend you all to add "\n" symbol after every sentence (or more complex formatting) for better reading.

[DatingWithWaddle]

For anyone getting this error FileNotFoundError: [WinError 2] referenced to audio.py or any error saying your audio file cannot be located, please make sure you have the ffmpeg executable in the same folder as your .py file. It will solve the issue. You're welcome.

[whynotanyting]

I know there's immense effort behind getting this to work, but the fact that us plebeians can use this tech in just a few lines is fucking magical.

[maxwellonyesoh]

Hi, for some reason my IDE isn't recognizing the whisper import although I've installed it

[sebastiangonzales]

Is it possible to use it using our own voice in realtime? like a speech to text

[krishna]

i am getting FileNotFoundError: [WinError 2] The system cannot find the file specified my file is in same directory Please naybody help me in this

[funTech_else_entrepreneurship]

Hey bro, I am trying to follow the steps to run this app on my local computer but it is giving me a run time error in installing whisper. Already tried googling it but nothing was found relevant-can I share screenshot with you?

[chandanabandaru]

Hey i am facing issue with the tensor when using whisper model for large audio files like 30 minutes ...can anybody help me how to resolve it?

[mustafamagdy]

thank you . but facing that issue , FileNotFoundError: [WinError 2] The system cannot find the file specified . the mp3 file in the same directory as the py script .

[malikshahid]

you should start a python course for beginners like me which know nothing about programming languages

[CanDoSo_org]

This is so cool! How can we transfer while recording? Also, is there a similar way to convert text to speech? Many thanks.

[gggyyy]

can it do witth time line SRT subtitles ?

[flyingzeppo]

When you said it took some time, and fast forwarded that part, approximately how long does it take to make the transcription on a laptop. I know "it depends", but are we talking approximately 2 minutes or more like 24 hours?

[RaphiAlmes]

Servas, daunk da recht narrisch. Prägnant und straighforward! Leiwaund

[Czarlsen]

What If I had two or more persons in the conversation. How can I distiniguish them

[asphoarkimete]

thanks i tried your tutorial it works perfectly and it is 10 times faster than speech recognite google...

i also tried to contact you on facebook, excuse me could you make a tutorial done well like yours, about how to build a neural network to clone a voice, very complex topic, but it will surely bring you many likes, because you can't find anything done well.

[pedroaugustobb]

How could I get the transcription if the audio is not in english?

[nachoagost]

where should i have installed ffmpeg. I already put the command pip install python-ffmpeg but it doesnt solve the error. it says No such file or directory: 'ffmpeg'