Okay, I tried something new with AI...

As you might know, I created a Free PowerShell Course (available on YouTube) - but it's only in German so far. I considered to use a 2nd Channel to post English content, but I don't want to divide the audience and the views into multiple channels. But I can nowadays upload multiple audio-tracks for a video (that was in beta for a few years until I can finally also use it). Since I didn't feel like dubbing >60 videos, I tried to use AI. YouTube itself does some auto-dubbing by default, but in my opinion that's complete garbage. Can't stand to watch any video like that, because e.g. the voices sound robotic and not like the original person. Background music and non-speech sounds are also lost.

So instead I used Qwen3-TTS to clone my own voice, translated the text via Google Gemini and created subtitles with OpenAI Whisper. I think the result is "good enough".

Initially this was way more work than if I would've just spoken the text myself, but I think in the long run this will save me a lot of time plus make my content more accessible. I could also add other languages, but I can't really judge the quality of the results. Next step is to automate the process :P

Please let me know what you think 😊

🎬 You can find the video here:

https://www.youtube.com/watch?v=h4y4qzJQg0w

#powershell #ai #qwen3

Okay, I tried something new with AI...

As you might know, I created a Free PowerShell Course (available on YouTube) - but it's only in German so far. I considered to use a 2nd Channel to post English content, but I don't want to divide the audience and the views into multiple channels. But I can nowadays upload multiple audio-tracks for a video (that was in beta for a few years until I can finally also use it). Since I didn't feel like dubbing >60 videos, I tried to use AI. YouTube itself does some auto-dubbing by default, but in my opinion that's complete garbage. Can't stand to watch any video like that, because e.g. the voices sound robotic and not like the original person. Background music and non-speech sounds are also lost.

So instead I used Qwen3-TTS to clone my own voice, translated the text via Google Gemini and created subtitles with OpenAI Whisper. I think the result is "good enough".

Initially this was way more work than if I would've just spoken the text myself, but I think in the long run this will save me a lot of time plus make my content more accessible. I could also add other languages, but I can't really judge the quality of the results. Next step is to automate the process :P

Please let me know what you think 😊

🎬 You can find the video here:

https://www.youtube.com/watch?v=h4y4qzJQg0w

#powershell #ai #qwen3