I publish multi-author books, and I want to create an audiobook, and the authors want to read their own chapters. I have 2 challenges: How do I combine all these audio files from different people into one audio file for the audiobook? Since these files come from different people, how do I maintain consistent quality so it sounds like it’s from one person? Is there a limit to the size of an audio file?
Thank you
Dina Marais
There’s a thing called EQ-matching which makes recordings more similar, but even then it will still be possible to tell they are different people recording in different locations, (in different rooms).
That’s the idea. But I think I have found a solution. AI Software that clones the author’s voice and then reads the chapter. Which is wonderful. Thank you for answering my question, Trevor.
Speech-to-speech AI retains the performance …
The “Loudness Normalization” effect can help all your narrators get on the same average loudness. I suggest setting each segment to -23 LUFS for audiobooks.
After that, using any of the many EQs included in Audacity (or ones you can download like TDR Nova), you can match the levels of treble, bass, mids, and other frequencies. This takes a bit of time and a keen ear, but you’ll be proud of the results and yourself as an editor once you get it down.
You can also use the Compressor and Limiter to solve issues like narrators moving closer to / away from the mic, and to bring up quieter passages of speech to an audible level.
You could also use a tiny bit of reverb to give the affectation of everyone being in a room, but less is more with reverb!