The Soundscape Perception Index (SPI)

In [1]:

import numpy as np
import pandas as pd

import soundscapy as sspy
from soundscapy.spi import MultiSkewNorm
from soundscapy.surveys.survey_utils import LANGUAGE_ANGLES, PAQ_IDS

data = sspy.isd.load()
data, excl_data = sspy.isd.validate(data)
data = data.query("Language != 'cmn'")

excl_id = [652, 706, 548, 550, 551, 553, 569, 580, 609, 618, 623, 636, 643]
data = data.drop(excl_id)

import numpy as np import pandas as pd import soundscapy as sspy from soundscapy.spi import MultiSkewNorm from soundscapy.surveys.survey_utils import LANGUAGE_ANGLES, PAQ_IDS data = sspy.isd.load() data, excl_data = sspy.isd.validate(data) data = data.query("Language != 'cmn'") excl_id = [652, 706, 548, 550, 551, 553, 569, 580, 609, 618, 623, 636, 643] data = data.drop(excl_id)

In [2]:

for lang in data.Language.unique():
    angles = LANGUAGE_ANGLES[lang]

    lang_idx = data.query(f"Language == '{lang}'").index
    iso_pl, iso_ev = sspy.surveys.processing.calculate_iso_coords(
        data.loc[lang_idx, PAQ_IDS], (1, 5), angles
    )
    data.loc[lang_idx, "ISOPleasant"] = round(iso_pl, 3)
    data.loc[lang_idx, "ISOEventful"] = round(iso_ev, 3)

data.head()

for lang in data.Language.unique(): angles = LANGUAGE_ANGLES[lang] lang_idx = data.query(f"Language == '{lang}'").index iso_pl, iso_ev = sspy.surveys.processing.calculate_iso_coords( data.loc[lang_idx, PAQ_IDS], (1, 5), angles ) data.loc[lang_idx, "ISOPleasant"] = round(iso_pl, 3) data.loc[lang_idx, "ISOEventful"] = round(iso_ev, 3) data.head()

Out[2]:

	LocationID	SessionID	GroupID	RecordID	start_time	end_time	latitude	longitude	Language	Survey_Version	...	THD_THD_Max	THD_Min_Max	THD_Max_Max	THD_L5_Max	THD_L10_Max	THD_L50_Max	THD_L90_Max	THD_L95_Max	ISOPleasant	ISOEventful
0	CarloV	CarloV2	2CV12	1434	2019-05-16 18:46:00	2019-05-16 18:56:00	37.17685	-3.590392	eng	engISO2018	...	-0.09	-11.76	54.18	34.82	26.53	5.57	-9.0	-10.29	0.209	-0.144
1	CarloV	CarloV2	2CV12	1435	2019-05-16 18:46:00	2019-05-16 18:56:00	37.17685	-3.590392	eng	engISO2018	...	-0.09	-11.76	54.18	34.82	26.53	5.57	-9.0	-10.29	-0.443	0.468
2	CarloV	CarloV2	2CV13	1430	2019-05-16 19:02:00	2019-05-16 19:12:00	37.17685	-3.590392	eng	engISO2018	...	-2.10	-19.32	72.52	32.33	24.52	0.25	-16.3	-17.33	0.637	0.002
3	CarloV	CarloV2	2CV13	1431	2019-05-16 19:02:00	2019-05-16 19:12:00	37.17685	-3.590392	eng	engISO2018	...	-2.10	-19.32	72.52	32.33	24.52	0.25	-16.3	-17.33	0.589	-0.092
4	CarloV	CarloV2	2CV13	1432	2019-05-16 19:02:00	2019-05-16 19:12:00	37.17685	-3.590392	eng	engISO2018	...	-2.10	-19.32	72.52	32.33	24.52	0.25	-16.3	-17.33	0.445	-0.126

5 rows × 144 columns

In [3]:

ct = data.query("LocationID == 'SanMarco'")
x = ct["ISOPleasant"].values
y = ct["ISOEventful"].values

msn = MultiSkewNorm()
msn.fit(data=ct[["ISOPleasant", "ISOEventful"]])
msn.summary()
tgt_df = pd.DataFrame(
    msn.sample(n=1000, return_sample=True), columns=["ISOPleasant", "ISOEventful"]
)
sspy.density(tgt_df, color="red")

ct = data.query("LocationID == 'SanMarco'") x = ct["ISOPleasant"].values y = ct["ISOEventful"].values msn = MultiSkewNorm() msn.fit(data=ct[["ISOPleasant", "ISOEventful"]]) msn.summary() tgt_df = pd.DataFrame( msn.sample(n=1000, return_sample=True), columns=["ISOPleasant", "ISOEventful"] ) sspy.density(tgt_df, color="red")

Fitted from data. n = 96
Direct Parameters:
xi:    [-0.106  0.559]
omega: [[ 0.149 -0.062]
 [-0.062  0.103]]
alpha: [ 0.697 -0.787]


Centred Parameters:
mean:  [0.1   0.381]
sigma: [[ 0.107 -0.025]
 [-0.025  0.071]]
skew:  [ 0.107 -0.128]

Out[3]:

<Axes: title={'center': 'Soundscape Density Plot'}, xlabel='$P_{ISO}$', ylabel='$E_{ISO}$'>

In [5]:

msn2 = MultiSkewNorm()
msn2.define_dp(
    xi=np.array([0.06534, 0.628637]),
    omega=np.array([[0.14890315, -0.06423752], [-0.06423752, 0.10139612]]),
    alpha=np.array([0.79105, -0.767217]),
)
msn2.sample(n=1000, return_sample=True)
msn2.summary()
tgt2_df = pd.DataFrame(
    msn2.sample(n=1000, return_sample=True), columns=["ISOPleasant", "ISOEventful"]
)
sspy.density(tgt2_df, color="purple")

msn2 = MultiSkewNorm() msn2.define_dp( xi=np.array([0.06534, 0.628637]), omega=np.array([[0.14890315, -0.06423752], [-0.06423752, 0.10139612]]), alpha=np.array([0.79105, -0.767217]), ) msn2.sample(n=1000, return_sample=True) msn2.summary() tgt2_df = pd.DataFrame( msn2.sample(n=1000, return_sample=True), columns=["ISOPleasant", "ISOEventful"] ) sspy.density(tgt2_df, color="purple")

Fitted from direct parameters.
Direct Parameters:
xi:    [0.065 0.629]
omega: [[ 0.149 -0.064]
 [-0.064  0.101]]
alpha: [ 0.791 -0.767]


Centred Parameters:
mean:  [0.283 0.451]
sigma: [[ 0.102 -0.026]
 [-0.026  0.07 ]]
skew:  [ 0.136 -0.131]

Out[5]:

<Axes: title={'center': 'Soundscape Density Plot'}, xlabel='$P_{ISO}$', ylabel='$E_{ISO}$'>

In [ ]: