Date   
Re: segmentation fault with long strings

Boersma Paul
 

At 18:38 +0000 23-11-06, psibre wrote:
If 28,018
or more are to be printed at once, the entire *line* containing the
28,018th character is not printed, and Praat crashes with a
segmentation fault.
This error in Praat was due to the fact that an (extended) line could not be longer than 28,000 characters. This and other mistakes have now been corrected in 4.5.06. The following tricks now work correctly with any length of strings, limited only by your computer's memory:

print 'text$'

text$ = Get label of interval... 1 1

a$ = "'b$'"

All three (the last one is strange, but still) used to crash with strings of a megabyte. They no longer do.

--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

Re: Relative addresses in scripts

Boersma Paul
 

At 20:07 +0100 12-12-06, Daniel Hirst wrote:
but now I have the scripts and man pages in separate folders I'm
stuck as to how to give the relative address for the man page from
the script
the only trick that works across all platforms is to put the man pages in a directory below the script directory. For instance, if you have the man pages in the directory "manual" inside the script directory, you do
Read from file... manual/Menu.man

If you want your plugin to run only on Windows XP, MacOS X and Linux (and not on Windows 98 or MacOS 9), you can put the man pages in a sister directory, as you did. You do
Read from file... ../manual/Menu.man
--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

Relative addresses in scripts

Daniel Hirst <daniel.hirst@...>
 

I'm working on a plugin for Praat.

I have a number of scripts and I also have a number of manual files.
I managed to get everything working with a sub-menu under New calling
the different scripts and another submenu under Help opening the man
files.

then I thought I would like to tidy up the plugin.

I would like to put the scripts inside a folder called "scripts" and
the man files in a folder called "man"

One of the scripts "plugin_manual.praat" reads the menu of the manual
and in the first version contained this

Read from file... Menu.man
but now I have the scripts and man pages in separate folders I'm
stuck as to how to give the relative address for the man page from
the script


any help would be more than welcome

------------------------------------------------------------------------
-------------------------
Equipe Prosodie et Représentation Formelle du Langage - salle A485
CNRS Laboratoire Parole et Langage (UMR 6057)
Université de Provence 29 avenue Schuman
13621 Aix-en-Provence cedex 1, France
tel: +33-4-4295-3628 secr: +33-4-4295-3634 fax: +33-4-4295-3788
------------------------------------------------------------------------
--------------------------
Speech Prosody list http://mailup.univ-mrs.fr/wws/info/speech-
prosody
Speech Prosody 2008 May 6-9 Campinas, Brazil (web site forthcoming)
Momel-Intsint list http://tech.groups.yahoo.com/group/momel-intsint
Praat-users list http://groups.yahoo.com/
group/praat-users
------------------------------------------------------------------------
--------------------------




[Non-text portions of this message have been removed]

Re: Reading a 32-bits sound file.

Boersma Paul
 

At 23:55 +0000 11-12-06, kyokonagao wrote:
Hello. I'd like to read 32-bit wav file, but Praat gives me an error
message 'unsupported windows audio encoding 3'.
Praat can read both 24-bit and 32-bit integer-encoded wav files, but your file seems to have a floating-point encoding. If you send a small example to me, I can make this type of files readable in Praat.

At 06:59 +0000 12-12-06, basu_a2001 wrote:
hi, I also faced a similar problem while reading 32 bit wav files
recorded for my experiments. PRAAT showed error messages. I had to
downsample it to 16 bits to open in PRAAT. Could anybody suggest us a
way to open 24/32 bit files so that we can retain maximum information
recorded in the data?
This incorrectly assumes that if Praat cannot read 32-bit files, then it cannot read 24-bit files either. Praat has been able to read 24-bit files (which are always integer-encoded) from May 2004 on, as well as 32-bit integer-encoded files.

--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

reading 32 bit wav files

basu_a2001
 

hi, I also faced a similar problem while reading 32 bit wav files
recorded for my experiments. PRAAT showed error messages. I had to
downsample it to 16 bits to open in PRAAT. Could anybody suggest us a
way to open 24/32 bit files so that we can retain maximum information
recorded in the data?

Amrita Basu

Reading a 32-bits sound file.

kyokonagao
 

Hello. I'd like to read 32-bit wav file, but Praat gives me an error
message 'unsupported windows audio encoding 3'. I tried both OSX
4.5.05 and Windows 4.5.04 versions, but had the same message. How can
I fix this problem?

Kyoko Nagao

intensity manipulation

gjohnsg <gabriel.murray@...>
 

I am wondering about the most efficient way to adjust the intensity of
various sound files, using IntensityTier. Say you have many wav files
that will differ in intensity, some very loud and some very quiet, and
you want them all to be relatively equal in intensity. This is what I
have at the moment. It adds one point to the tier, and the intensity
is 70 dB minus the mean intensity. Is this an efficient way to do this?



select Sound one
To Intensity... 100 0 yes
intmean = Get mean... 0 0 energy

# we want all audio files around 70dB

diff = 70-intmean
tmax = Get end time

Create IntensityTier... empty 0 1
Add point... tmax diff
select Sound one
plus IntensityTier empty
Multiply... no
Play

Re: increasing naturalness of vowel synthesis

Boersma Paul
 

At 11:21 +0000 11-12-06, Jyrki Tuomainen wrote:
I still get this attack-like onset
did you already try lowering the "adaptation factor" to 0.5 or so? If not, then try; if you did, then a multiplication with an IntensityTier may help.

the vowel sounds monotonous
this should depend on not much more than the PitchTier that you used to create the voice pulses. You could include a short rise at the beginning. Other suggestions are welcome.

and the voice is "tight".
a spectral issue. Make sure that you apply at least 10 formants. I assume that you did that all right, so now I'm waiting for suggestions.

how to add some jitter
Select the PointProcess, choose "To Matrix", then "Formula":
self + randomGauss (0, 0.0001)
then "To PointProcess". Perhaps this is a good moment for me to make a Formula command for PointProcesses.

what would be good values for
those parameters that are available in the To Sound (phonation) command?
Powers of 2 and 3 (rather than 3 and 4) would yield something close to Rosenberg's glottal source model, which was also used in the Klatt synthesizer. But 3 and 4 have been chosen as defaults because they seem to approximate better the known relations between human glottal parameters.

Any better wisdom on this would indeed be very welcome.
--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

something odd with SSCP ? (or maybe it's just me ...)

cgendrot@univ-paris3.fr
 

Dear Praat users, Dear Paul

I am having a problem with the following lines (that I've been using for a long time though)
I intend to draw formant (F1 and F2) ellipses from data contained in the first 2 columns of a text file.

select 'table_temp'
To SSCP... 0 0 1 2
sscp = selected ("SSCP")
Draw confidence ellipse... 'niveau_confiance' 2 1 'en2$' 'en1$' 'enonces$' 'en3$' no

... and seems very odd to me ...
if the number of rows of the table is less than 46342, then everything is OK and I get my ellipse right
If it is more, then no ellipse appears at all
But then again, if the number of rows is more than 66000 (approximately this time), then it appears back.

In this last case, it seems that the confidence level gets smaller and smaller as the ellipse gets progressivel smaller.
But I progressiveley increased my data from 46341 to 46342 by artificially copying old data lines (thus not increasing standard deviation), and I could see the ellipse suddenly disappear.
Is there something I am doing wrong ?

Thanks for any hints ....


cedric
Laboratory of Phonetics and Phonology, Paris

increasing naturalness of vowel synthesis

Jyrki Tuomainen
 

Hello list.
I've created a source signal from scratch using PitchTier and PointProcess for vowel synthesis.

"To Sound (phonation)" command provides a couple of parameters to manipulate the naturalness of the glottal source, and I've had some success, but would still like to increase the naturalness. My problem is that I still get this attack like onset, and the vowel sound monotonous and the voice is "tight". I think I've got the pitch contour alright, but does anyone have suggestions for what other parameters I could try (e.g. how to add some jitter), and/or what would be good values for those parameters that are available in the To Sound (phonation) command?


Thanks, and best wishes, -Jyrki

--
New temporary address from July 2006 to Autumn 2007

Jyrki Tuomainen, University College London, Human Communication Science
Remax House, 31/32 Alfred Place, London WC1E 7DP, UK
Tel: +44 (0)20 7679 4214 (Internal 24214), Fax: +44 (0)207 679 4238
j.tuomainen@..., http://www.hcs.ucl.ac.uk/
--

Superposition of voices question

prosodylover
 

I have to study a lot of conversations in which there are some
superpositions of voices. I'm new in Praat, and i was asking myself if
there is any possible to split the voices of any superposed segment in
a serie of individual files.
I'll try to explain me: I have two persons talking at the same time
and i want to get the individual sound of each one separately, with
the less interference or noise from the other sound source. Is this
possible? I don't know with Praat, but maybe you know some method or
some program that could do this (perhaps from music world or forensic
acoustic?). Please, my faculty thesis begin here. Help me.
Many thanks. By the way, i admire you: P.Boersma and D.Weenink. This
is an amazing program. I have some suggestions about navegation on it,
but i don't want to be disrespectful. I just want to colaborate.
Have a very nice day, my phonetic and phonologic friends!

Re: Pitch/amplitude-table?

Boersma Paul
 

At 18:53 +0100 9-12-06, Daniel Hirst wrote:
mySound = selected("Sound")
To Pitch... 0.01 75 600
To Matrix
myPitchMatrix = selected("Matrix")
select mySound
To Intensity... 70 0.01 yes
Down to Matrix
plus myPitchMatrix
Merge (append rows)
Transpose
This is not guaranteed to give the same number of measurement points ("frames") for the Pitch and the Intensity. Typically, the Intensity will have fewer frames than the Pitch.

It is notoriously difficult to get your analyses to yield measurement points at the same times. This is because the unmeasurable parts of the sound (at the beginning and end) have different a duration for each analysis.

So the only way is to create the two analyses with a small time step, then create a list of interpolated points:

mySound = selected ("Sound")
tmin = Get start time
tmax = Get end time
To Pitch... 0.001 75 600
myPitch = selected ("Pitch")
select mySound
To Intensity... 70 0.001 yes
myIntensity = selected ("Intensity")
time = ceiling (tmin / 0.01)
echo Time F0 intensity
while time < tmax
select myPitch
f0 = Get value at time... time Hertz Linear
select myIntensity
intensity = Get value at time... time Cubic
printline 'time:6' 'f0:3' 'intensity:3'
time += 0.01
endwhile

This gives you a complete list in the Info window. You will see "--undefined--" values for pitch and intensity at the edges, and some more "--undefined--" values for pitch where the sound is voiceless.
--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

Re: "Hint: file name starts with a space"

Boersma Paul
 

At 03:09 +0000 8-12-06, gaboriccho wrote:
Write to WAV file... 'file$'

And got the error:
"Cannot create file: " D:&#92;mySound.wav" "
"Hint: file name starts with a space"
yes. Your line will be expanded as follows, if 'file$' contains "D:&#92;mySound.wav":

Write to WAV file... D:&#92;mySound.wav

So you typed two spaces after the three dots. The first space comes with the dots, so that the second space will be regarded as the space with which the file name begins. If your intended file name does not begin with a space, don't type it.

After half an hour of frustration (and useless searching) I realized
that I had put TWO SPACES in:
Write to WAV file...<SPC><SPC>'file$'
Tut-tut. Bad, bad style!

I am really looking forward to use versions where this not exactly
use-friendly feature (together with some other white space issues) is
reconsidered.
This is impossible. If I removed this feature, you could never open files whose names start with a space any longer.
--

Paul Boersma
Phonetic Sciences, University of Amsterdam
Spuistraat 210, room 303
1012VT Amsterdam, The Netherlands
http://www.fon.hum.uva.nl/paul/
phone +31-20-5252385

Re: Pitch/amplitude-table?

Daniel Hirst <daniel.hirst@...>
 

The reason that you have a different number of values for pitch and
amplitude is because you used the default ("0.0 = auto") value for
the time step. If you do "To Pitch..." and "To Intensity..." with
the same time step (e.g. 0.01) then you will get Pitch and Intensity
objects with the same number of values.

You can then convert these to Matrix form, select them both and do
Merge and then transpose the resulting Matrix which you can then
saveas a headerless spreadsheet file which will have pitch and
intensity values on each line for every 10ms (ie 0.01s) of the sound.


the following script will do this assuming the Sound object is selected:

==================================================

mySound = selected("Sound")
To Pitch... 0.01 75 600
To Matrix
myPitchMatrix = selected("Matrix")
select mySound
To Intensity... 70 0.01 yes
Down to Matrix
plus myPitchMatrix
Merge (append rows)
Transpose
==================================================

there may well be a simpler solution but this seems to work!

good luck!

daniel
------------------------------------------------------------------------
-------------------------
Equipe Prosodie et Représentation Formelle du Langage - salle A485
CNRS Laboratoire Parole et Langage (UMR 6057)
Université de Provence 29 avenue Schuman
13621 Aix-en-Provence cedex 1, France
tel: +33-4-4295-3628 secr: +33-4-4295-3634 fax: +33-4-4295-3788
------------------------------------------------------------------------
--------------------------
Speech Prosody list http://mailup.univ-mrs.fr/wws/info/speech-
prosody
Speech Prosody 2008 May 6-9 Campinas, Brazil (web site forthcoming)
Momel-Intsint list http://tech.groups.yahoo.com/group/momel-intsint
Praat-users list http://groups.yahoo.com/
group/praat-users
------------------------------------------------------------------------
--------------------------

Re: Pitch/amplitude-table?

Ingmar Steiner
 

Hello,

if I understand your question correctly, you have two TableOfReals,
one with a column for time and another for pitch values, the other
with a column for time and another for amplitude values. Also, the
tables do not necessarily have the same number of rows. You want to
merge the two tables, so that you have one column for time, a second
column with the pitch values from the first table, and a third column
with the amplitude values from the second table. Also, you want to
have the time points in the first column in ascending order, i.e. you
want to "interleave" two tables along the values of a common column.
Assuming this is what you want, here's how you could do it:

Select both TableOfReals and use the "Append" command from the
Synthesize menu. However, before you do this, you have to modify your
two original tables so that the pitch and amplitude values are
retained as separate columns in the merged table. This means that you
have to insert padding ("dummy") columns into your original tables, so
that the number of columns in your tables is the same, but e.g. in the
pitch table, there is an empty third column, while in the amplitue
table, there is an empty column between the time and amplitude
columns. The time values, on the other hand, should be in the same
column in both tables (I'm assuming column 1).

Once you have prepared your tables in such a way, the "Append" command
will simply do what its name implies. All time values are in column
1, all pitch values are in column 2, and all amplitude values are in
column 3. All you have to do now is sort the table, so that the time
values are in order, which you can accomplish with the "Sort by
column..." command.

I don't know if this is what you had in mind. Also, it might be a
problem that all data in the padding cells is set to 0. You may want
to interpolate (or something similar) in a subsequent step. But I hope
I could help.

Best wishes,

-Ingmar Steiner
_______________
Phonetics/Phonology Section
Department of Computational Linguistics and Phonetics
Saarland University

Office: Building C7 2, Room 5.06
Email: steiner@...
Phone: +49-681-302-4696
http://www.coli.uni-saarland.de/~steiner/

--- In praat-users@..., "doctorbu0" <bastufloejt@...> wrote:

Forgive me for being so stupid
I suppose this program is very beautiful and intelligent
but
what i simply want to do
is to create a table with
pitch and amplitude along the same timesteps.

I managed to make a table (of real) with 2 columns (pitch & time)
and
another one with amplitude & time
problem is

1. I do not know how to merge the two.
2. The amp-table has only half as much timesteps as the pitch-table
although in their tier receptively they seem to be the same.

I´m not a programming person and the Table Help as well as the
Amplitude
Tier Help seems not to be written yet.

I use the Mac OS 9 version...

thank you anyone

/Bu

Re: Create DurationTier from script

Ingmar Steiner
 

Hello,

the error in your script is due to the fact that you're using "select"
instead of "plus" before trying to replace the DurationTier. The
"Replace DurationTier" command is available only when a Manipulation
AND a DurationTier object are selected (as hinted at in Praat's error
message).

Note that your script also lacks the actual resynthesis command, e.g.
"Get resynthesis (PSOLA)", before which you would have to deselect the
DurationTier. Otherwise, you would get another error, because the
command "Play" is not avalailable for Manipulation objects, only "Play
(PSOLA)" and "Play (LPC)" (and I'm assuming you want to create a
lengthened Sound).

However, for what you seem to want to to, the following "script" is
quicker:

###
Lengthen (PSOLA)... 75 600 1.5
Play
###

Best wishes,

-Ingmar Steiner
_______________
Phonetics/Phonology Section
Department of Computational Linguistics and Phonetics
Saarland University

Office: Building C7 2, Room 5.06
Email: steiner@...
Phone: +49-681-302-4696
http://www.coli.uni-saarland.de/~steiner/

--- In praat-users@..., "gjohnsg" <gabriel.murray@...>
wrote:

Hello, I have been using Praat for a while, but primarily through the
user interface. I am wanting to change the duration and intensity of
some wav files and am wondering how to do this entirely from script.
As a simple example, say I want to create a new duration tier and add
a single point where the audio slows down by a factor of 1.5. This is
what I wrote:

#####
To Manipulation... 0.01 75 600
Rename... manip
Create DurationTier... shorten 0 0.9
Add point... 0.5 1.5
select Manipulation manip
Replace Duration Tier
Play
#####

However, I get the response "Replace Duration Tier not available for
current selection". Could someone point out where I am going wrong here?
Many thanks

Re: Festival catalogue file to TextGrid

bloit <Julien.Bloit@...>
 

Hi Rattima,

I guess you could use one of the scripts on Mietta's Praat scripts (http://www.helsinki.fi/~lennes/praat-scripts/).

If anyone's is interested, I have also written two python scripts which perform the 2 ways conversion between a set of N text files and a textgrid file with N tiers. I'm new to this list and I don't know if I can upload these files in the "files" section of this group, without messing up the existing page. Can someone tell me if there's there a special moderation process for uploading?

Hi,

I'm trying to convert the Festival catalogue file into TexGrid, so I can have a graphic display of where the boundaries are in the wave form. I wonder if anyone has a script for doing this. Here's the format of the Festival catalogue file:

dh_0 nancy_001 0.065000 0.077500 0.090000
ax_0 nancy_001 0.090000 0.129000 0.168000

It has phoneme labels, starting point, midpoint, and endpoint. And the following is the output I'm looking for after the conversion:

intervals [1]:
xmin = 0 xmax = 0.065000
text = "dh_0" intervals [2]:
xmin = 0.065000
xmax = 0.090000
text = "ax_0"
Also, another script for converting this back to the catalogue file format will be very helpful.

Thank you very much,
Rattima



To Post a message, send it to: praat-users@...
To Unsubscribe, send a blank message to: praat-users-unsubscribe@...
To consult archives : http://egroups.com/list/praat-users/
Yahoo! Groups Links




"Hint: file name starts with a space"

gaboriccho <gaboriccho@...>
 

Hi,

I just write to save some time for those who had the same problem.
In a script a wanted to write my selected Sound into a .wav file.

[...]
file$="D:&#92;mySound.wav"
Write to WAV file... 'file$'
[...]

And got the error:
"Cannot create file: " D:&#92;mySound.wav" "
"Hint: file name starts with a space"

Was there any space in my variables? (in principle I dont use spaces
in files)
After half an hour of frustration (and useless searching) I realized
that I had put TWO SPACES in:
Write to WAV file...<SPC><SPC>'file$'
Tut-tut. Bad, bad style!

After deleting the extra space it worked fine.

I like Praat a lot, it is a great tool.
And I am really looking forward to use versions where this not exactly
use-friendly feature (together with some other white space issues) is
reconsidered.


Gabor

Pitch/amplitude-table?

doctorbu0
 

Forgive me for being so stupid
I suppose this program is very beautiful and intelligent
but
what i simply want to do
is to create a table with
pitch and amplitude along the same timesteps.

I managed to make a table (of real) with 2 columns (pitch & time)
and
another one with amplitude & time
problem is

1. I do not know how to merge the two.
2. The amp-table has only half as much timesteps as the pitch-table
although in their tier receptively they seem to be the same.

I´m not a programming person and the Table Help as well as the Amplitude
Tier Help seems not to be written yet.

I use the Mac OS 9 version...

thank you anyone

/Bu

Create DurationTier from script

gjohnsg <gabriel.murray@...>
 

Hello, I have been using Praat for a while, but primarily through the
user interface. I am wanting to change the duration and intensity of
some wav files and am wondering how to do this entirely from script.
As a simple example, say I want to create a new duration tier and add
a single point where the audio slows down by a factor of 1.5. This is
what I wrote:

#####
To Manipulation... 0.01 75 600
Rename... manip
Create DurationTier... shorten 0 0.9
Add point... 0.5 1.5
select Manipulation manip
Replace Duration Tier
Play
#####

However, I get the response "Replace Duration Tier not available for
current selection". Could someone point out where I am going wrong here?
Many thanks