fix tests

tts/text/input_parsing/parser.py

@@ -8,7 +8,7 @@ Inspired by and including code from original Tacotron2 by Keith Ito.
 # pyright: strict
 
 import re
-from typing import List, Union
+from typing import List, Union, Any
 
 # Regular expression matching text enclosed in curly braces:
 _curly_re = re.compile(r'(.*?)\{(.+?)\}(.*)')
@@ -16,10 +16,14 @@ _curly_re = re.compile(r'(.*?)\{(.+?)\}(.*)')
 class ArpabetSequence:
     def __init__(self, sequence: str):
         self.sequence = sequence
+    def __eq__(self, other: Any) -> bool:
+        return isinstance(other, ArpabetSequence) and self.sequence == other.sequence
 
 class GraphemeSequence:
     def __init__(self, sequence: str):
         self.sequence = sequence
+    def __eq__(self, other: Any) -> bool:
+        return isinstance(other, GraphemeSequence) and self.sequence == other.sequence
 
 ParsedSequence = List[Union[ArpabetSequence, GraphemeSequence]]
 

tts/text/phonetization/test_english_phonetization.py

@@ -1,5 +1,7 @@
 from tts.text.phonetization.cmudict import DEFAULT_CMUDICT_PATH, CMUDict
 from tts.text.phonetization.english_phonetization import EnglishPhonetization
+from tts.text.input_parsing.parser import ArpabetSequence
+from tts.text.input_parsing.parser import GraphemeSequence
 
 class TestEnglishPhonetization:
 
@@ -13,20 +15,59 @@ class TestEnglishPhonetization:
         phoneme_dictionary = CMUDict(DEFAULT_CMUDICT_PATH)
         phonetization = EnglishPhonetization(phoneme_dictionary)
         output = phonetization.try_convert_to_phonemes("this is a test")
-        assert output == ["DH IH1 S", "IH1 Z", "AH0", "T EH1 S T"]
+        assert output == [
+            ArpabetSequence("DH IH1 S"), 
+            ArpabetSequence("IH1 Z"),
+            ArpabetSequence("AH0"), 
+            ArpabetSequence("T EH1 S T"),
+        ]
 
     def test_most_words_in_dictionary(self):
         phoneme_dictionary = CMUDict(DEFAULT_CMUDICT_PATH)
         phonetization = EnglishPhonetization(phoneme_dictionary)
         output = phonetization.try_convert_to_phonemes("nonword is a test")
-        assert output == ["nonword", "IH1 Z", "AH0", "T EH1 S T"]
+        assert output == [
+            GraphemeSequence("nonword"), 
+            ArpabetSequence("IH1 Z"), 
+            ArpabetSequence("AH0"), 
+            ArpabetSequence("T EH1 S T"),
+        ]
 
     def test_punctuation(self):
         phoneme_dictionary = CMUDict(DEFAULT_CMUDICT_PATH)
         phonetization = EnglishPhonetization(phoneme_dictionary)
         output = phonetization.try_convert_to_phonemes("this...")
-        assert output == ["DH IH1 S", "..."]
+        assert output == [
+            ArpabetSequence("DH IH1 S"), 
+            GraphemeSequence("..."),
+        ]
         output = phonetization.try_convert_to_phonemes("this!")
-        assert output == ["DH IH1 S", "!"]
+        assert output == [
+            ArpabetSequence("DH IH1 S"), 
+            GraphemeSequence("!"),
+        ]
         output = phonetization.try_convert_to_phonemes("this?")
-        assert output == ["DH IH1 S", "?"]
+        assert output == [
+            ArpabetSequence("DH IH1 S"), 
+            GraphemeSequence("?"),
+        ]
+
+    def test_extra_spaces(self):
+        phoneme_dictionary = CMUDict(DEFAULT_CMUDICT_PATH)
+        phonetization = EnglishPhonetization(phoneme_dictionary)
+        output = phonetization.try_convert_to_phonemes("    this    is    a     test   ")
+        assert output == [
+            ArpabetSequence("DH IH1 S"), 
+            ArpabetSequence("IH1 Z"),
+            ArpabetSequence("AH0"), 
+            ArpabetSequence("T EH1 S T"),
+        ]
+
+    def test_multi_non_word_extra_spaces(self):
+        phoneme_dictionary = CMUDict(DEFAULT_CMUDICT_PATH)
+        phonetization = EnglishPhonetization(phoneme_dictionary)
+        output = phonetization.try_convert_to_phonemes("    nonword    anothernonword   ")
+        assert output == [
+            GraphemeSequence("nonword"), 
+            GraphemeSequence("anothernonword"), 
+        ]