How can you improve voice and diction quality?

Post your questions & get help from friendly LibriVoxers
Post Reply
Posts: 60
Joined: January 8th, 2021, 2:10 pm

Post by KimL » January 27th, 2021, 1:51 pm

I am a newbie to volunteering for Librivox. I've completed 3 sections from the short works forum. In listening to my recordings, I feel like I would like to improve my voice, diction quality. Here are the areas I am interested in getting tips for.

1. I feel like my voice is a little nasally. I have been running my recordings through the "filter curve" and boosting the bass. I feel like that helps a little but it is just something I am analyzing by listening. Is there a technical way to reduce nasality? Are there any tips for using the filter curve? Should I not be using it?

2. I am trying to figure out which words or sounds to put stress on and where to put pauses to give some animation to the reading. Are there any suggestions for this or is it just a matter of artist interpretation and figuring out how to use your voice?

3. I sometimes feel like my tempo starts out good and then speeds up in the middle (maybe when i am getting fatigued). Should I try to prevent this? Should i take breaks and not try to do too much in one sitting. I fear if I take a break, my voice might sound different in the second segment.

Any other tips people can give would be much appreciated.


Posts: 2970
Joined: January 17th, 2013, 9:16 pm
Location: Rochester, NY

Post by commonsparrow3 » January 27th, 2021, 3:05 pm

Hello, Kim!

I listened to a little bit of your voice, and it doesn't sound particularly nasal to me at all. Actually, it sounds like a familiar western NY accent! I'm a native of Rochester NY, and I've got the region's famous accent with all those weird nasal vowels. (My hometown of Rochester is pronounced something like "Rat-chaster" by us natives.) Occasionally, I've encountered people not from these parts who roll their eyes at my accent. But I just tell myself "That's what I sound like", and I don't worry about it. I think your voice sounds fine, and you shouldn't worry about it either. (I have no idea what a "filter curve" is, I must admit. My audio tech skills are very basic.)

As to your second question, deciding where to put the emphasis as you read is not going to have any "right" answer. It's up to you -- you are the voice artist! You choose how to interpret the material. When I read, it's very common for me to re-read a phrase more than once, with different emphasis, trying out which way I think best pulls out the meaning. When I edit, I can sit back and listen to how the various readings sound, and pick the one that strikes me as best. But that's just how I do it, not necessarily the only way.

As to your last question, I know what you mean about your voice starting out one way, then ending up another. With me, I usually initially "attack" in much too quick a voice, then slow down once I get more "into" the material. I've found that reading the intro and first paragraph more than once helps to ease me into my regular cadence. I delete the "warm-up" paragraph from the beginning when I edit. If I can't read the whole section in one sitting, I find it helps me get back into my previous voice if I listen to the last minute or two of the old recording while reading aloud along with it, trying to match the tone and cadence. It may not end up matching perfectly, but usually it's close enough not to be jarring. It also helps if I record the second part at the same time of day as the first part, as my morning voice and my evening voice are rather different, and make it harder to get a "match".

I hope you are enjoying yourself at LV! We were all "newbies" once ourselves, and it's always a delight to welcome yet more newcomers to the LV family!

Posts: 60
Joined: January 8th, 2021, 2:10 pm

Post by KimL » January 27th, 2021, 4:25 pm

Wow, Maria! You have given great advice. Thank you. I'm going to try your tips and tricks with my next recording. I found the info very valuable.

Its funny you sensed a New York accent with my reading. Guess what! I'm originally from New Orleans!!! I have heard people refer to the New Orleans and NY accents as "port accents" so perhaps that is a relation. I now live in Michigan and have lived here for 20-something years. So, throw in midwestern and I guess that is what I have. ha ha. :D

I am enjoying this new endeavor with LV. Its fun. I feel like I'm dipping my toe into acting a little bit, but I don't have to be self conscious on a stage. I also really enjoy reading the little pieces that I have read. Once you read a piece a ton of times, you really start to grasp and remember the story. I've had a few neat stories so far.

Thanks again for your help. Happy reading to you!!!


Peter Why
Posts: 4959
Joined: November 24th, 2005, 3:54 am
Location: Chigwell (North-East London, U.K.)

Post by Peter Why » January 28th, 2021, 12:42 am

With regard to your question about emphasis, I try to read something as I would like to hear it. Think about how it would sound if you heard it in a drama on the radio or television. It can help, especially for shorter pieces, not to read it "cold" (reading cold is reading without going through it before recording); so you would read it through with the microphone off, and try out different ways of emphasis, possibly marking up your script to remind yourself. Ham it up a little.

It'll come to you ... and don't worry when you eventually get someone criticising your voice. We all do (I've seen comments about my "nasal" readings) ... and you'll get compliments, too. Remember the compliments, ignore the criticism (or, if it does worry you, ask for a second opinion, here on the forum).

"I think, therefore I am, I think." Solomon Cohen, in Terry Pratchett's Dodger

Posts: 26
Joined: May 27th, 2016, 10:26 am
Location: Mid-Michigan

Post by audiomike » January 28th, 2021, 8:13 am

Question 1: If the nasal thing really bothers you, try the following. Open the Filter Curve effect and start with a straight line. Place a marker on the line at 1,000 Hz, another about 200 Hz to the left, and the last about 200 Hz to the right. Now drag the center marker all the way to the bottom of the graph. This creates a high Q notch filter at 1 kHz and can be effective in reducing nasally voices. You may have to play around with the center (1,000) frequency to find what works best for your voice.

Questions 2 & 3: Many people have a problem when starting out that they don't even realize. They seem to think that reading into a microphone means they have to change the way they speak. It doesn't. What they wind up doing is speaking "at" the microphone instead of "to" it. You have to try to forget the fact that there's a microphone in front of you. When you read, imagine yourself sitting around a camp fire telling the story to a group of people. This will help you to read naturally as if that microphone was a person listening to you. When it's not natural, it shows. The more conscious effort you exert while reading, the worse it will sound. Learn to relax and let the story flow without "trying" to get it right. Remember, you're not "reading" the story, you're "telling" it. Like everything else worth doing, it takes time, practice, and patience.

Hope this helps.

Posts: 5
Joined: August 4th, 2009, 1:29 pm

Post by ChristopherW » January 28th, 2021, 10:29 am

I've read only part of it myself, but The Speaking Voice seems like good reading (or listening).

Posts: 60
Joined: January 8th, 2021, 2:10 pm

Post by KimL » January 28th, 2021, 3:04 pm

I appreciate all these tips! I'm going to try the trick with the Filter Curve from Audiomike.

Thanks again!!!

Posts: 1780
Joined: December 6th, 2010, 5:15 pm

Post by realisticspeakers » January 28th, 2021, 3:21 pm

Hi there!

One thing I was taught and still struggle with is to imagine you are speaking into the ear of someone who really wants to hear the story you are telling.
When you are intimately speaking into someone's ear, your voice lowers in pitch, and then your breathing 'should' become easier thus bringing the voice from a lower source (nasality).
Imagine your voice emanating from your belly.

I "prepare" the text--make marks, write notes, add new lines to break up phrases, and get really really familiar with the context to try and deliver it more smoothly than a cold read.
That helps in where to put emphasis, as the text becomes "mine".
Someone once said to only allow one word to be emphasized per sentence. It's not really a rule of thumb but a general guideline when you're not sure.
It is surprising how much better you sound to yourself when you "read flat" because emphasizing too many words at random doesn't sound "real".
Truth exists for the wise, Beauty for a feeling heart: They belong to each other. - Beethoven
"Kind reader, if this our performance doth in aught fall short of promise, blame not our good intent, but our unperfect wit."

LibriVox Admin Team
Posts: 34105
Joined: April 3rd, 2008, 3:55 am
Location: Melbourne,Australia

Post by annise » January 28th, 2021, 4:02 pm

Saying much the same, early on I was told to imagine reading it to someone sitting or lying near you and it worked for me :D . And it is the way the readers I prefer read
But everyone is different - some readers treat it as a performance and some listeners may prefer that. And that's OK too.


Post Reply