Hello everybody
I make Youtube videos and I would like the sound of my voice to be dynamic and punchy.
Usually I do a 20db noise reduction, then a boost, then a limiter and finally an EQ, but I find the sound a bit flat and monotonous.
What equalizer setting or other function could I use to get a dynamic sound. Or is there simply a plugin that could do this automatically?
True dynamics come from the performance… Sing the quiet parts quietly and the loud parts loudly and you have more dynamic range or “dynamic contrast”.
If you over-boost and “try” to push the peak over 0dB you get clipping which is dynamic compression plus distortion (it reduces dynamics).
You probably want dynamic compression or limiting to give your voice more “loudness” or “intensity” Try the limiter effect (with make-up gain to bring-up the loudness) and if you want to add some “grit” (distortion) set it for clipping. Virtually all modern commercial recordings have compression/limiting except for maybe some classical & jazz recordings.
Compression/limiting on the vocal can help them stand-out above the music in a mix.
NOTE - As you may know the peaks are limited digitally to 0dB. But our perception of loudness mostly depends on the short-term average level. More dynamic music has a lower average (it’s “less dense”) so it tends to sound quieter even with the same peaks. That’s no problem if you have enough amplification one analog side because you can just turn it up. But compression and limiting win the Loudness Wars.
Thank you for your reply.
I may have misspoken but I don’t do music, I do Youtube videos, facecam videos and voiceovers.
I do a boost but never beyond 0db
That’s the musical version of “Acting.” You want the Theatrical Acting Macro (TAM). As soon as someone writes it, we’ll contact you.
I think I may have uncovered the next April First posting. “Tired of ordinary, flat vocal presentation?”
Actually, there is a way around this. Given the current YouTube postings, I’m betting you can get an AI presenter to announce your work for you. That’s your Google Search Term. Post back if you find something.
Koz
I think the “vocal expression” you’re looking for is more related to pitch & “tone” variations. Try to mimic or copy the style of an announcer you like…
And I would still try limiting for more-constant “intensity”. And it wouldn’t hurt to try the limiter settings with clipping.
That might be doable with something like Autotune to give the voice performance some variation.
The problem is it can’t be just any variation. Pitch and volume get managed to enforce meaning or emphasize a point. One of the ways you can tell that a language isn’t native is the speaker gets the volumes wrong.
Another problem with “fake” generation that I actually heard on-line is the software used the same inflections again…and again…and again… It worked OK the first time…
That’s not to say that pitch, volume, and variations can’t be used to create a pleasant, believable presentation, it’s just that it would be a retirement project. How old are you?
Koz
And to bring this around, you could learn how to announce. Depending on your setup, live monitoring of your live voice in sealed headphones (very highly recommended) can go a very long way to help you create an entertaining product. Most people are horrified the first time they hear themselves. “Oh my goodness. That’s not me, is it?”
I would also reference the audiobook people who arrive on the forum with a basket of effects, filters, modifications, and apps. Nice, but you are going to have to apply that whole basket to all your chapters, plus editing and corrections. See: Retirement Project.
Koz
Of course. My suggestion to use Autotune was more or less tongue-in-cheek. It can automatically vary pitch to make it more “dynamic”, but it won’t necessarily sound good.
I developed a simple text-to-speech synthesizer, so I know of some of the challenges in getting automatic inflections to sound right. But I only worked with primary/secondary stresses on syllables within a word (e.g. “,in-con-'cei-va-ble”) and simple things like inflecting the last word(s) of a question (e.g., “How about a nice game of chess?” with “chess” being stressed). More than that would require the text-to-speech converter to understand English sentence structure (parts of speech etc.) and emotional context, which is probably beyond the capabilities of the microcontroller I developed the synthesizer for.
Maybe AI can do it? 
See: Me, about half-way up this message.
If you do get one to work, post back how you did it.
Koz
Is that from a speech synthesizer? It’s sometimes hard to tell these days with how good they’ve gotten.
Or Trebor in the broom closet with a Yeti.
Koz
That’s more common than you think. There is a YouTube presenter (that I don’t follow any more) who actually did a sponsor announcement and got the name of the product wrong.
Koz
This topic was automatically closed after 30 days. New replies are no longer allowed.