commit to clean up

acoustic_model/check_novoapi.py

@@ -25,12 +25,12 @@ mapping = convert_xsampa2ipa.load_converter('xsampa', 'ipa', default.ipa_xsampa_
 
 stimmen_transcription_ = pd.ExcelFile(default.stimmen_transcription_xlsx)
 
-phonelist_novo70_      = pd.ExcelFile(default.phonelist_novo70_xlsx)
-df = pd.read_excel(phonelist_novo70_, 'list')
-
 
 ## novo phoneset
-#translation_key = dict()
+translation_key = dict()
+
+#phonelist_novo70_      = pd.ExcelFile(default.phonelist_novo70_xlsx)
+#df = pd.read_excel(phonelist_novo70_, 'list')
 ## *_simple includes columns which has only one phone in.
 #for ipa, novo70 in zip(df['IPA_simple'], df['novo70_simple']):
 #    if not pd.isnull(ipa):
@@ -38,3 +38,18 @@ df = pd.read_excel(phonelist_novo70_, 'list')
 #        translation_key[ipa] = novo70
 #phonelist_novo70 = np.unique(list(df['novo70_simple']))
 
+phoneset_ipa = []
+phoneset_novo70 = []
+with open(default.cmu69_phoneset, "rt", encoding="utf-8") as fin:
+	lines = fin.read()
+	lines = lines.split('\n')
+	for line in lines:
+		words = line.split('\t')
+		if len(words) > 1:
+			novo70 = words[0]
+			ipa	   = words[1]
+			phoneset_ipa.append(ipa)
+			phoneset_novo70.append(novo70)
+			translation_key[ipa] = novo70
+phoneset_ipa    = np.unique(phoneset_ipa)
+phoneset_novo70 = np.unique(phonset_novo70)