From 8efb091715c651a0f716b8f5029a44545584114c Mon Sep 17 00:00:00 2001
From: yemaozi88 <428968@gmail.com>
Date: Tue, 15 Jan 2019 11:30:49 +0100
Subject: [PATCH] confusion matrix is output.

---
 .vs/acoustic_model/v15/.suo                   | Bin 71168 -> 76800 bytes
 .../__pycache__/defaultfiles.cpython-36.pyc   | Bin 1099 -> 1167 bytes
 acoustic_model/check_novoapi.py               |  48 +++++++++++++-----
 acoustic_model/defaultfiles.py                |   1 +
 4 files changed, 36 insertions(+), 13 deletions(-)

diff --git a/.vs/acoustic_model/v15/.suo b/.vs/acoustic_model/v15/.suo
index f5fd7886def8bd17ad70bdd8dbb5b128a60a9480..08c3df7ab00772ca32e7c35f2553440b21d0c278 100644
GIT binary patch
delta 2931
zcmcJRe@s+Y7RTScJceI_1B}S808;Eu`B7%*3=G4FGs6Qx!TEtwYm^@%h!qv-h*oJG
ztWCRF+XYUuQ}>UJKGSYu8jJIaR%?`wq-mO#mZnWj<Bv9Ly0Nt>nC-@Dgq{O~(ny;o
zO>gqyoO9o~_uczF_uaw!hUmR0_Ud4)IU-Rl2;qVtgdh`>lapu!hRi?Vy2Dr&EaK@g
zV_&>_;p`iqv6b|24ea52*i!hBPpbA}9J2^+ha~S?fG!T7YN_aFAyG&Kl7Y-a79w{s
z<0-TyNHh|MBqJ$EB+`jkkXYmqBoPr2nwZv?i+&DbN0uYhS1t&BwXxAL7*bjST1}rT
zebJ&|hdF5JA+;oommrxiE-qYx0^D19%`BfXAHq0(R$mqR4#XAgtA+AVhxjx`@M@^m
zlpgFZMC(Dig7Kc9r7535zXj<=yhs`Xx7ezn*3KK3)TFf)yMLi|E5@XCSmxq9bZIIh
z+8K>0rbZf*tp{6p3R@5B#fAHH*f*mcJtOVN&NMy7ixH~L!)QrJEixq}+I*x4c@$A0
zq<=9&bCSj;gfuQdN|9`2Mq^B=n_P+=B%KW@LktMXE+G{NEo2$mN@NA{Sg@}etuyFT
z4mEc$u7j$GhxF?(BGpOp2BZ;bLP+81{OP|`qD@G9pDP{)BEr=>W+m)IAKs5hO{Ln*
z9!0g;hx8!*ksdW|IGs&z{9&|kAaYeKb(%0JM$AYF*rHyDUXNZq(tu3#H9glq&%;V#
z=e&Jv3H*qEi(qH;bK#dr&Od4L(91MA=;PuQb=+8(KSSaoRT;d-T*Z@lkonwk%&J(^
zhFHJWsbm!lF#pQ6EY`dy?MPVcf-_q+d9SIJqagxG_N-&S@6#A&R?J!7Fv!T&iu7r~
z%m3DbC)W|dNVUwFp8w@KVK`BD)Kkou1$LT8-~!i!rv&pz(Z2-qqW+4`F$9}$G=`NZ
z=Ir;FY~kQd%~UcUy%Hjqa@ipIt+zPKP=2z9F_5Cvv>PUs;_X5A1>OS9<!x)5ciC*x
zQ#Cf5tw^#pZ|vH!%!PhsOYLg6y-sR#cGT7EZql!GS2&AWHmt3#ZKsiNFknzin^%{u
zSYdy1&BxQz!6yM-xM{0Pa@Dl8^<U5?)m3*G+H<z(a@xzCc4PjoD(BWBcU@WW5cCwj
z1esS2P<$XOx!GP^*<@eo?s8QcS_^a9H`SE8cP-n!+*u^oaarR2b(|}_qJoQS>CIEZ
zRXRu*RSF)Fa{}DX)q=bxGyDmBPB$avgs#J9-$0)+9}XK+mEkH2f3S{n0q@VP{2c`W
zPu@W)dI7vQ5u;dFVd2f5->OjL{R$})k8{y)xuu$_gd_A1Yy+e!%k(mm$1)W6lgF97
zoXgv||6fulog47f9+bO8xthz}T(0M`m&=V@J|M~)MA^xKcUFM6MXl7e9%Jyzj<=?Q
z;SNiP{0jHGC3bHP4~LAFMy2uD!%VJMfxR<9N!b4_lQ(ktWiGe!Uk>i&G7cF1J|FV?
z&-i*pmZ-3=&x=v?8PRV57XtA2Gk=4zyJ|S}rWOwU{{m=_=PTZK7=wW`OOFnn<zPLP
zLkVJc_g)iB-hLjQzmqP82@(xWf}!(PXgX;HmKN{J)-a8-X()-o5v3qF=_Y7A6*c9j
z_+Lu6h%cRP1w+@-p{rGTcWMn^4bF%5EgI;(qXEa@f1>G=5*o)BG>i|P&r`DB78zW9
zJ1J6U)|vO}^@SS?z-CB<wBa@wI&Jmo;#eXi1~x=&7S;*$V=TatUFooNby`^PEAX9&
zXX(%y(5T8x#-hja@H>E>lZompr?vLu6`uvzwWJRM`s>wTJC%*=Rmq>KFhy%KNCoA_
zB3<D#>gKkV<}D4|ws&oB>e&3iuoKsF?``uMEr6PLdYe{s=iKIeW>VPjB!e2myx?9m
z%8MWFX8zvmd{)8eQoUa=qdX!)(`t>P`SJz>Hkbedcb#(<P#HRFW%4BsogXGZ!ntFA
zu$aYg`M;khjMC1gn*O(oxCZn+T19*1B7+m(IiTb%%L7#n{gtX};pR9V^Bad59^GUm
zc?_@Uf$yrIVXzW@7%Y2W9y}d~0x94<?fnyu!mkx+t`k@f*G{L>8dBlh$SS%^BP_yp
zp#y)uw}ETir8sX6u>Of}SpY^qqPw;*l>QI-C-hvZkw-YVN3&q$YB-o38E{~-`=rQB
zA<((lsW?9wU^7Jz@tMNd`%phFoswc0SNST`?B;#3zx*f;N-tL{^*7{u@`cR(&K<K~
W_PWT#`;)>yjM;caXUvI?hyM)Hix_MG

delta 3160
zcmdUxdrZ_<7RTq_`3(XB@>YSufIOyZ<>62Ui#WjWbjr-gqe9z7L<I$5lo<=CIJ7ls
zt3ig-F3q-S%+F1C)9$+M{MsV57DQs2?v_>KAI&y#W73$_HAW0gw!5pd-#g5>8oQg#
zru)ZU@|kn)xv$?n_ndoieJ#0uC5_~ya+SYCs#IZ6)!?4j>lId^FZ<B;H^!1Fm8V3E
zG@X6**!!1R9qg_K8sG!0fxhG3AQuc3TKcOxp*kNbhWm;M`qN-B2zkK16neS8{uJaA
za2F#IAfrJjI6%(_FLghc{@6iO{PTGupsxn;AQqImZA&xapgaM}z&a4<EaY<3*MnM`
zla{W}gjNJ3unIibb0g|jAoebb1$qszfwiET$~6{Aw9uOx-TFN+4S-(Y0Q-Pw?DtEd
zAA-IgtOQSjy+GLS(OqVNOD_g#!+-wkq7isB8%+;+v=s*7(M4$)kNJnCLk2vlg`Nt8
zC&iTER1%zc=qIQNKSn`54sfKZM34e9Kr#pdV(&U2_L>Em23o)xkOUraGi=o$2jqf0
zPyocr!t+AH^NlJ4DutjJ1Y9o~H=u3=C7=T6fjA7&7ATddR{=9v@j%-qoH2WjszX-f
zZh`rE5O8QCbg_r+kb!B0+<|&K=yKa)=!1}8=5*T<>2gS<yFNB}$lrN1)Dfv=a1IcY
z2<YX|`a=aU7(gLd2c7~&U_B@X8-Nj%fKpHfOrRW8fJ#sWszD9d2+Tk%Pzz}RR$%kn
z0*m`4c|-TO?S|1X2dgJ%*dbO&-{JRZa)!U8F^CkUyCE<9|9$!z->2tB$0(7}w-=E<
zI#MW1PT*ss63ZJi$FK;+m^{Ti2_`OykR(_BYQ>Rsl#zVEK*m4DE4AsD7-^HzA8sDi
zS~YK=|4)@v>-$Vz$wxD*#8`4<TId|lrTI*&BtmMzm~~WMBgs}S>!p9!@q{Cbu}$RE
z&1#rPo`u1W`(|AHDh%%d6vg}O_dzUE`F^WKBQ?Q5!{a(~R9Qk84%{2=w`dY^+_%V`
zpH0W}lW3;82vX}Khf50TT(<MEJ1NXW?_5czj<h@JYQxpK;`l2q+rs8so^3weCYBR(
ziqfx9slLIKG`QkL&t8){SX>VxnHQFK9`$+kjq}2(^;1kXa(T1lF>$W&x9@4>*Ek(N
z)~GZNzsE*(j{jtzV{VD9rGz_LWz(*cOkU09ZZ3~V@|#@VC&_tSKEUOlaoNRXnaeLr
z9>ZU_&nt`6^1Iw~={l!7yOt4Cyz=YZQ&GkLYhMZ9$scjKhI{NL%_IAk2f62yDHbZ?
zEe8IORMaw&O20W3C%+?!eGi@0k!>OymQFf$IEva&WvN3{CV#K{Ho7snK_2F`=Xbx1
zS1`{)(M^2?1-@6A&eN6&t#6hiLveI&GAll|yJufdt9@t74rh0pV`ooy?;3l*nb<7@
z9T|#c_ce0Qmqm%oHc-U9sZy@qNMTexe<WUno2qdxu4V4yNpV+;_tEs#{pDPTBgoPd
zA(mu`RML}(D?+TXU=M$|Ce{iUkK5mUVUsoWrE+y%Vt8iNFp9D+2Y>tZCMwY{qvY8R
zGR_!BPYtt2_D92~Rw}Lu#whCSh7(hta%w%8N4q;CM$fqD;^cA`_{1QZI%71dRDVKz
z_Tu6H7Se&IYZ|K^#Ia05?nYuv^MA&3hiz}6-eTFl&0xr1o874|7$BA$qtt!=1f#j}
z?SajaYi1z2uQj)JlXyo|&E~qU_SP;3nTqw1n|~iFY$``{X+yK4CXZez&Y`Jdovyyk
zR_N?2vp1UU`mR--Ijg$Lt)=<dFI<_>m7c7)Jk7pc{^fM;bu+#7c}9z~t+Ks;*XFW<
ztpimC{dTJ(x2nH)YgKk-$1^>(y^evBTuV<~MMtx<&$g$c-O#(ksjqN4O17=4H*Bmc
zZ`xh3ub{Ffr>Waq*45r*ZY^os)lgWgZ|o~HZEH1HN-1SBHLN+eu&>SF*qN8#P@u3`
zFQdF;Gs?_Jks9el(X$SmL9Fu1g+4~NKN(a8|L~%k(%g2%ettwFpM>8VV#i}c*-bTV
z`6h@8hSC(>&950<oU5keH<M`A7(|_(L6L9@QzoUw;@Mcpg}@QTi8NS(b78<&b{Ye1
zoqlKgBJ77#<NCGu(-<Y5Ec^>G^;_uAfd5%eSmfKBtkXq5c4{P1lqX?4A%R^}(*;x3
z4<t%(AXpwwkH!amgv2<vkS^-AW#uQ3B6Mrq=*y6(TRP-^96zw1yq_JWm(FN$&m{fG
z`N95d$4wUbI=eCMNn!D9EH9SrpldU+$URl)VsX>Zx#A4<lLyb@5{U~ABno{vXAlWO
om)>0#`A9NYPk+D`qVpGDXqogc!qRiWG#<8u@u2aD73@>?59k>*C;$Ke

diff --git a/acoustic_model/__pycache__/defaultfiles.cpython-36.pyc b/acoustic_model/__pycache__/defaultfiles.cpython-36.pyc
index 635aaaed776cc3945d38c00f15e5d070810f80f6..9a6baf33b5bc4669f059ecfc577c096caf0d778f 100644
GIT binary patch
delta 433
zcmZvY&q~8U5XQ5cttM&ywN2HwiiarJiwG4wi9UfBZ;KE@W~(V}Qj_Y{Lk=EAgU=x1
z3+NMg@C|bHs_)<wmLlRVJM(=%hL2%Cv^(7&SeEhjcz8B76y;O-b(T;c<IBJ9{i7S9
zjOa)eHPApSRIY)}b;)!vpbloixFLxFRzZ1v;cRp!INXxN0vCxboed3i4m7zdi32Sp
zjm5MFgEpwJQq*`;LLF9N4Hs!i-+~Twp$F?Y;{iEU`Wq!9m;K5erdcE)2*>exG7-s)
zx5FsH9z^3X&*Nbng|j$K%3I=-s(c~ePPLq**J(J7xf<pfFQ4cpX_xQxWWVbDO$#O|
zT#ZElamG9U2k)w{UGB^y5B7rTMVg3Q%$}9Z#Lvyl2e!Slg-T<LxgH}nUY9b`oGa{D
L8KXAws7rqUpZsqz

delta 346
zcmYk2y-LJD5XbjR-F$6wd-3Z|5ey1;f{2wUK7ftgA{@b4^$tzW<kZq44<NEnVPoTC
z2o{#X&RU<r6_&$+VdnRne>IQZ8_#y4$iMIJo^l_dFZ8cw?S5#x{(Q;~9)K&Z(TI$w
z=HP)3LBo{ij0a&uls~1&QUF4Q&V&$K6FG_?u_U0QVrK+MYm#ZY3n@^TZ)Q}-xs1aC
zEH+e0CnPMvGW4KtLtVfpXluo!_(rPaysXYN4DzetMRBc*iFv^pR^|hL$7-%9@5+2U
zY-wIst$8PFI5j`y*xvW)s=1}Fy*B>K>}?Ommt~=AJ$Xh|VBb&$=7?=<uG^I}#*&m_
Rfdye$x@n0qB5+C)@(cU&PEP;;

diff --git a/acoustic_model/check_novoapi.py b/acoustic_model/check_novoapi.py
index 1814a6b..aa6c993 100644
--- a/acoustic_model/check_novoapi.py
+++ b/acoustic_model/check_novoapi.py
@@ -3,23 +3,27 @@ os.chdir(r'C:\Users\Aki\source\repos\acoustic_model\acoustic_model')
 
 import sys
 import csv
-#import subprocess
-#from collections import Counter
-#import re
 
 import numpy as np
 import pandas as pd
-#import matplotlib.pyplot as plt
-#from sklearn.metrics import confusion_matrix
+import matplotlib.pyplot as plt
+
+from sklearn.metrics import confusion_matrix
+from sklearn.metrics import accuracy_score
+import novoapi 
+
+from forced_alignment import pyhtk, convert_phone_set
 
 import acoustic_model_functions as am_func
 import convert_xsampa2ipa
 import defaultfiles as default
-
-from forced_alignment import pyhtk, convert_phone_set
-
-import novoapi 
 import novoapi_functions
+sys.path.append(default.accent_classification_dir)
+import output_confusion_matrix
+
+## procedure 
+forced_alignment_novo70 = True
+
 
 ## ===== load novo phoneset =====
 phoneset_ipa, phoneset_novo70, translation_key_ipa2novo70, translation_key_novo702ipa = novoapi_functions.load_phonset()
@@ -145,9 +149,9 @@ for word in word_list:
 
 
 ## ===== forced alignment =====
-if forced_alignment:
+if forced_alignment_novo70:
 	Results = pd.DataFrame(index=[],
-		columns=['filename', 'ipa', 'word', 'result_ipa', 'result_novo70', 'llh'])
+		columns=['filename', 'word', 'ipa', 'result_ipa', 'result_novo70', 'llh'])
 	for word in word_list:
 	#for word in ['Oor']:
 		# pronunciation variants top 3
@@ -176,7 +180,7 @@ if forced_alignment:
 		samples = df_.query("ipa in @pronunciation_ipa")
 
 		results = pd.DataFrame(index=[],
-			columns=['filename', 'ipa', 'word', 'result_ipa', 'result_novo70', 'llh'])
+			columns=['filename', 'word', 'ipa', 'result_ipa', 'result_novo70', 'llh'])
 	
 		#j = 0
 		for i in range(0, len(samples)):
@@ -205,5 +209,23 @@ if forced_alignment:
 		Results.to_excel(os.path.join(default.stimmen_dir, 'Results.xlsx'), encoding="utf-8")
 else:
 	Results_xlsx = pd.ExcelFile(os.path.join(default.stimmen_dir, 'Results.xlsx'), encoding="utf-8")
-	R = pd.read_excel(Results_xlsx, 'Sheet1')
+	Results = pd.read_excel(Results_xlsx, 'Sheet1')
 
+
+## ===== analysis =====
+result_novoapi_dir = os.path.join(default.stimmen_dir, 'result', 'novoapi')
+for word in word_list:
+	if not word == 'Oog':
+#word = 'Reus'
+
+		Results_ = Results[Results['word'] == word]
+		y_true  = list(Results_['ipa'])
+		y_pred_ = [ipa.replace(' ', '') for ipa in list(Results_['result_ipa'])]
+		y_pred  = [ipa.replace('ː', ':') for ipa in y_pred_]
+		pronunciation_variants = list(set(y_true))
+		cm = confusion_matrix(y_true, y_pred, labels=pronunciation_variants)
+
+		plt.figure()
+		output_confusion_matrix.plot_confusion_matrix(cm, pronunciation_variants, normalize=False)
+		#plt.show()
+		plt.savefig(os.path.join(result_novoapi_dir, word + '.png'))
\ No newline at end of file
diff --git a/acoustic_model/defaultfiles.py b/acoustic_model/defaultfiles.py
index e9c4c96..b9ab0ab 100644
--- a/acoustic_model/defaultfiles.py
+++ b/acoustic_model/defaultfiles.py
@@ -29,6 +29,7 @@ config_hvite = os.path.join(cygwin_dir, 'config', 'config.HVite')
 repo_dir = r'C:\Users\Aki\source\repos'
 ipa_xsampa_converter_dir    = os.path.join(repo_dir, 'ipa-xsama-converter')
 forced_alignment_module_dir = os.path.join(repo_dir, 'forced_alignment')
+accent_classification_dir   = os.path.join(repo_dir, 'accent_classification', 'accent_classification')
 
 WSL_dir   = r'C:\OneDrive\WSL'
 fame_dir        = os.path.join(WSL_dir, 'kaldi-trunk', 'egs', 'fame')