Parallel Annotation of Speech and Text

This page is under construction

Project Description

Goal of this short pilot has been parallel sound and text annotation. The study has been conducted by Professor Wim van Dommelen and Assc.Professor Dorothee Beermannat the Institute of Languages and Communication Studies at the Norwegian University of Science and Technology.Scientific assistant for the project was Asger Hagerup. The project has been funded by the SSTL.

The pilot investigated integrated presentations of linguistically annotated audio and text material, combining Praat and TypeCraft.

Praat is a signal analysis software developed by Paul Boersma and David Weenink from the University of Amsterdam. It is a tool widely used for the annotation of sound objects. For the present study we have taken advantage of the fact that Praat annotation data resides in a TextGrid object that exists separately from the sound object.Specifying a sentence tier allowed us easy referencing of data across applications. At present our sound signal representations are static, and selective, that is, they focus on the presentation of one selected feature to illustrate interesting correlations across phonetic and linguistic categories.

Description of the material

For our study we selected 10 sentences from the phonetic database of the Sound to Sense project.

Sentences 1 to 3

Speaker dialect: Bergen

Jeg ser bildet, kan du si, litt på skrått ned, ovenifra.

“I see the picture, say, somewhat diagonally downwards, from above.”

Jeg

e

1SG

PN

ser
se:	r
see	PRES
V

bildet
bild	e
picture	DEFSG
N

kan

kan:

canPRES

V

du

ʉ

2SG

CL

si

si:

sayINF

V

litt

lit:

a.little

ADVm

på

po

onDIR

PREP

skrått
skro:	t
diagonal	ADJ>ADV
ADVm

ned

ned

downDIR

ADVm

ovenifra

ovenifra

from.aboveDIRSRC

ADVm

Download files for viewing in the Praat application: File:PSTA01.mp3, TextGrid

Det dekker omtrent hele det venstre…mest…altså, venstreste kortsiden.

“It covers approximately the whole left…most…that is, the leftest short side.”

Det

de

3SGNEUT

PN

dekker
dek:	er
cover	PRES
V

omtrent

umtrent

approximately

ADVm

hele
he:l	e
whole	DEF
ADJ

det

de

DEFSGNEUT

ART

venstre

venstre

left

ADVm

mest

mest

mostSUP

ADJ

altså

aso

that.isDM

ADVm

venstreste
venstre	st	e
left	SUPMU	DEF
ADJ

kortsiden
kort	sid	en
short	side	DEFSG
N

File downloads for Praat: * Sound, * TextGrid # for help with the download

Hun står med ryggen mot veggen opp og ser på han som skal kaste ballen som står utenfor og peker på boksene.

“She's standing with her back up against the wall and looking at him, who is standing outside and about to throw the ball, and pointing towards the boxes.”

Hun

hun

3SGFEM

PN

står
sto:	r
stand	PRES
V

med

med

withMNR

PREP

ryggen
ryɡ:	en
back	DEFSG
N

mot

mut

againstDIR

PREP

veggen
veɡ:	en
wall	DEFSG
N

opp

up

upDIRMU

PREP

og

o

and

CONJC

ser
se:	r
see	PRES
V

på

po

atDIR

PREP

han

han

3SGMASC

PN

som

som

PNrel

skal

skal:

shallPRES

V

kaste
kast	e
throw	INF
V

ballen
bal:	en
ball	DEFSG
N

som

som

PNrel

står
sto:	r
stand	PRES
V

utenfor

ʉtenfor

outside

ADVm

og

o

and

CONJC

peker
pe:k	er
point	PRES
V

på

po

atDIR

PREP

boksene
boks	ene
box	DEFPL
N

Download files for viewing in the Praat application:Sound, TextGrid

Speaker Dialect: Trondheim

Parallel Processing of Speech and Text Data - Part 2

Speaker Dialect:

Parallel Processing of Speech and Text Data - Part 3

About the TextGrid files

The TextGrid files are opened together with the matching sound files for viewing in the Praat application. The TextGrid files consist of three tiers, 'Word' (rendered in Bokmål orthography) 'Phoneme' (shows underlying segments) and 'Note' (shows surface realisation with IPA symbols, and other notes).

Here is a list of glosses used in the 'Note' tier:

Phonology/Phonetics:
BrV = Segent realised with breathy voice
CrV = Segent realised with creaky voice
DV = Underlying voiced segment realised devoiced
EPN = Epenthesis
RD = Reduction of segment (e.g. corner vowel realised as schwa or plosive as fricative).
V = Underlying non-voiced segment realised voiced

Morphophonology/Syntax
CL = Clitic

Other
ERR = The speaker errs and corrects himself
HES = (Audible) hesitation from speaker