Linguistics Forum

Reply to topic

 
View previous topic :: View next topic
Author Message
zaba
Chomsky


Joined: 03 Apr 2010
Posts: 813
PostPosted: Tue Aug 03, 2010 7:04 am
Post subject: Measuring F1 & F2 in vowels with Praat
Reply with quote

What's the best way to measure F1 & F2 of a bunch of vowels? As I see it there are two approaches in Praat:

(1) highlight the vowel and get the average F1 and F2
(2) Put the cursor in the middle of the vowel and get the F1 & F2.

Which method is best?
Back to top
View user's profileSend private message
polarfrog
Good Linguist


Joined: 02 May 2010
Posts: 31
Location: Germany
PostPosted: Wed Aug 04, 2010 7:04 am
Post subject:
Reply with quote

do it by hand Razz

praat isn't often very precise, esp if you have not too great recordings etc. when you do it by hand you can check for influence from the surrounding sounds etc...
Back to top
View user's profileSend private message
zaba
Chomsky


Joined: 03 Apr 2010
Posts: 813
PostPosted: Fri Aug 06, 2010 6:49 am
Post subject: Yup, sounds about right!
Reply with quote

Thanks
Back to top
View user's profileSend private message
anabelle345



Joined: 30 Sep 2011
Posts: 1
PostPosted: Fri Sep 30, 2011 6:13 am
Post subject:
Reply with quote

Hello

I guess 1750 Hz is not a normal F2-value for [u] and [U] uttered by women.
We would expect the values to be 1.2 of the male values (ca. 800 Hz).

I often found problems also with buy cialis now male voices, e.g. measuring 2100 Hz for the F2 of [u].
As far as I remember, this is said to be a general problem of the LPC-algorithm which for back vowels either lumps together F1 and F2 or simply does not "see" F2, so often the indicated F2 is actually F3.

Unfortunately, I can't remember which parameters need to be adjusted (number of filters? window length?) nor how to do it with Praat.
Hopefully, somebody more experienced in acoustic phonetics can help.

Regards

Last edited by anabelle345 on Tue Dec 06, 2011 1:26 am; edited 1 time in total
Back to top
View user's profileSend private message
joice129



Joined: 20 Oct 2011
Posts: 1
PostPosted: Thu Oct 20, 2011 12:47 pm
Post subject:
Reply with quote

For synthesizing a single phoneme of a vowel, what kind of "activation versus time" I need for the required components? Is a constant activation enough for each component or it must change over time? If it needs to change, what sort of change? linear (start and end points) or triangular (start, middle and end points) or ...?
Back to top
View user's profileSend private messageVisit poster's website
Display posts from previous:   

All times are GMT - 5 Hours

Reply to topic

Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum