annotations.Intsint package

Submodules

annotations.Intsint.intsint module

filename

sppas.src.annotations.Intsint.intsint.py

author

Tatsuya Watanabe, Brigitte Bigi

contact

develop@sppas.org

summary

Algorithm of INTSINT automatic annotation.

class annotations.Intsint.intsint.Intsint[source]

Bases: object

Provide optimal INTSINT coding for anchor points.

DOWN = 0.25
HIGHER = 0.5
LOWER = 0.5
MAX_F0 = 600
MAX_RANGE = 2.5
MEAN_SHIFT = 50
MIN_F0 = 60
MIN_PAUSE = 0.5
MIN_RANGE = 0.5
STEP_RANGE = 0.1
STEP_SHIFT = 1
TONES = ['T', 'M', 'B', 'H', 'L', 'U', 'D', 'S']
TONES_ABSOLUTE = ['T', 'M', 'B']
TONES_RELATIVE = ['H', 'L', 'U', 'D', 'S']
UP = 0.25
__init__()[source]

Create a new Intsint instance.

adjust_f0(f0)[source]

Return F0 value within self range of values.

Parameters

f0 – (float) Input pitch value.

Returns

(float) Normalized pitch value.

annotate(momel_anchors)[source]

Provide optimal INTSINT coding for sequence of target points.

Parameters

momel_anchors – (list of tuple) List of time (in seconds)

and anchors (Hz).

estimate(tone, last_anchor)[source]

Estimate f0 from current tone and last target.

Parameters

  • tone –

  • last_anchor –

init(momel_anchors)[source]

Initialize INTSINT attributes from a list of targets.

Parameters

momel_anchors – (list of tuple) List of time

(in seconds) and anchors (Hz).

optimise(mid, _range)[source]

Fix tones.

Parameters

  • mid –

  • _range –

recode()[source]

Recode within the parameters space.

mean +/- 50 Hz for key and [0.5..2.5 octaves] for range.

reset()[source]

Fix all member to their initial value.

annotations.Intsint.intsint.linear(value)[source]
annotations.Intsint.intsint.octave(value)[source]

annotations.Intsint.sppasintsint module

filename

sppas.src.annotations.Intsint.sppasintsint.py

author

Brigitte Bigi

contact

develop@sppas.org

summary

SPPAS integration of the INTSINT automatic annotation.

class annotations.Intsint.sppasintsint.sppasIntsint(log=None)[source]

Bases: annotations.baseannot.sppasBaseAnnotation

SPPAS integration of the INTSINT automatic annotation.

__init__(log=None)[source]

Create a new instance.

Parameters

log – (sppasLog) Human-readable logs.

static get_input_extensions()[source]

Extensions that the annotation expects for its input filename.

INTSINT requires momel anchors which can either be stored in a TextGrid file or in a PitchTier file.

Returns

(list of list)

get_input_tier(input_files)[source]

Return the tier with Momel anchors.

Parameters

input_files – (list)

Raise

NoTierInputError

Returns

(sppasTier) Tier of type Point

get_output_pattern()[source]

Pattern this annotation uses in an output filename.

run(input_files, output=None)[source]

Run the automatic annotation process on an input.

Parameters

  • input_files – (list of str) momel anchors

  • output – (str) the output file name

Returns

(sppasTranscription)

static tier_to_anchors(momel_tier)[source]

Initialize INTSINT attributes from a Tier with anchors.

Parameters

momel_tier – (sppasTier) A PointTier with float values.

Returns

List of tuples (time, f0 value)

static tones_to_tier(tones, anchors_tier)[source]

Convert the INTSINT result into a tier.

Parameters

  • tones – (list)

  • anchors_tier – (sppasTier)

Module contents

filename

sppas.src.annotations.Intsint.__init__.py

author

Brigitte Bigi

contact

develop@sppas.org

summary

INTSINT automatic annotation.

INTSINT is an acronym for INternational Transcription System for INTonation. It was originally developed by Daniel Hirst in his 1987 thesis as a prosodic equivalent of the International Phonetic Alphabet, and the INTSINT alphabet was subsequently used in Hirst & Di Cristo (eds) 1998 in just over half of the chapters.

INTSINT codes the intonation of an utterance by means of an alphabet of 8 discrete symbols constituting a surface phonological representation of the intonation:

T (Top), H (Higher), U (Upstepped), S (Same), M (mid), D (Downstepped), L (Lower), B (Bottom).

These tonal symbols are considered phonological in that they represent discrete categories and surface since each tonal symbol corresponds to a directly observable property of the speech signal.

INTSINT is computed from a set of selected F0 anchors. The implementation into SPPAS corresponds to the most recent version of the algorithm:

De Looze, Céline & Hirst, Daniel. (2010).
Integrating changes of register into automatic intonation analysis.
Proceedings of the Fifth International Conference on Speech Prosody,
Chicago
class annotations.Intsint.Intsint[source]

Bases: object

Provide optimal INTSINT coding for anchor points.

DOWN = 0.25
HIGHER = 0.5
LOWER = 0.5
MAX_F0 = 600
MAX_RANGE = 2.5
MEAN_SHIFT = 50
MIN_F0 = 60
MIN_PAUSE = 0.5
MIN_RANGE = 0.5
STEP_RANGE = 0.1
STEP_SHIFT = 1
TONES = ['T', 'M', 'B', 'H', 'L', 'U', 'D', 'S']
TONES_ABSOLUTE = ['T', 'M', 'B']
TONES_RELATIVE = ['H', 'L', 'U', 'D', 'S']
UP = 0.25
__init__()[source]

Create a new Intsint instance.

adjust_f0(f0)[source]

Return F0 value within self range of values.

Parameters

f0 – (float) Input pitch value.

Returns

(float) Normalized pitch value.

annotate(momel_anchors)[source]

Provide optimal INTSINT coding for sequence of target points.

Parameters

momel_anchors – (list of tuple) List of time (in seconds)

and anchors (Hz).

estimate(tone, last_anchor)[source]

Estimate f0 from current tone and last target.

Parameters

  • tone –

  • last_anchor –

init(momel_anchors)[source]

Initialize INTSINT attributes from a list of targets.

Parameters

momel_anchors – (list of tuple) List of time

(in seconds) and anchors (Hz).

optimise(mid, _range)[source]

Fix tones.

Parameters

  • mid –

  • _range –

recode()[source]

Recode within the parameters space.

mean +/- 50 Hz for key and [0.5..2.5 octaves] for range.

reset()[source]

Fix all member to their initial value.

class annotations.Intsint.sppasIntsint(log=None)[source]

Bases: annotations.baseannot.sppasBaseAnnotation

SPPAS integration of the INTSINT automatic annotation.

__init__(log=None)[source]

Create a new instance.

Parameters

log – (sppasLog) Human-readable logs.

static get_input_extensions()[source]

Extensions that the annotation expects for its input filename.

INTSINT requires momel anchors which can either be stored in a TextGrid file or in a PitchTier file.

Returns

(list of list)

get_input_tier(input_files)[source]

Return the tier with Momel anchors.

Parameters

input_files – (list)

Raise

NoTierInputError

Returns

(sppasTier) Tier of type Point

get_output_pattern()[source]

Pattern this annotation uses in an output filename.

run(input_files, output=None)[source]

Run the automatic annotation process on an input.

Parameters

  • input_files – (list of str) momel anchors

  • output – (str) the output file name

Returns

(sppasTranscription)

static tier_to_anchors(momel_tier)[source]

Initialize INTSINT attributes from a Tier with anchors.

Parameters

momel_tier – (sppasTier) A PointTier with float values.

Returns

List of tuples (time, f0 value)

static tones_to_tier(tones, anchors_tier)[source]

Convert the INTSINT result into a tier.

Parameters

  • tones – (list)

  • anchors_tier – (sppasTier)