Peter Jackson's documentary ‘Get Back’ anticipated what became impressively audible to everyone with the release of the Beatles' last single “Now and then”: with the help of artificial intelligence, quantum leaps in the processing of sound and language can be achieved today.

What is particularly impressive is that numerous tools based on these new technologies are accessible to everyone and can be used in practice - in many cases even free of charge. Services such as Whisper from OpenAI or the German platform DeepL can be used to create transcripts and translations, improve text quality and generate subtitles and summaries of longer texts. The Austrian provider Auphonic can be used to enhance poor audio tracks and automate previously complex processes with just a few lines of script code.

Using a practical example, I will show you how to use artificial intelligence and a few lines of PowerShell code to optimise the sound of existing video material.

- optimised in terms of sound: Adjust volume, remove noise, enhance the sound of the voice (levelling, normalisation, voice EQ, denoising)
- created a transcript of the spoken word
- a summary of the content is generated

In German, with English subtitels: Live from CIM Lingen 2024
thorsten-butz.de/cimlingen2024/