Fix P, V, and E sounding odd. Add Slash to the punctuation list

Also add torch and torchaudio back to the requirements, as silero needs them. Silero's requirements.txt should be everything needed to run the tests

extensions/silero_tts/tts_preprocessor.py

@@ -2,13 +2,13 @@ import re
 
 from num2words import num2words
 
-
+punctuation = r'[\s,.?!/)"\'\]>]'
 alphabet_map = {
     "A": " Ei ",
     "B": " Bee ",
     "C": " See ",
     "D": " Dee ",
-    "E": " Ii ",
+    "E": " Eee ",
     "F": " Eff ",
     "G": " Jee ",
     "H": " Eich ",
@@ -19,13 +19,13 @@ alphabet_map = {
     "M": " Emm ",
     "N": " Enn ",
     "O": " Ohh ",
-    "P": " Pii ",
+    "P": " Pee ",
     "Q": " Queue ",
     "R": " Are ",
     "S": " Ess ",
     "T": " Tee ",
     "U": " You ",
-    "V": " Vii ",
+    "V": " Vee ",
     "W": " Double You ",
     "X": " Ex ",
     "Y": " Why ",
@@ -55,7 +55,7 @@ def preprocess(string):
 
     # cleanup whitespaces
     # remove whitespace before punctuation
-    string = re.sub(r'\s+([,.?!\'])', r'\1', string)
+    string = re.sub(rf'\s+({punctuation})', r'\1', string)
     string = string.strip()
     # compact whitespace
     string = ' '.join(string.split())
@@ -71,13 +71,13 @@ def remove_surrounded_chars(string):
 
 def replace_negative(string):
     # handles situations like -5. -5 would become negative 5, which would then be expanded to negative five
-    return re.sub(r'(\s)(-)(\d+)([\s,.?!)"\'\]>])', r'\1negative \3\4', string)
+    return re.sub(rf'(\s)(-)(\d+)({punctuation})', r'\1negative \3\4', string)
 
 
 def replace_roman(string):
     # find a string of roman numerals.
     # Only 2 or more, to avoid capturing I and single character abbreviations, like names
-    pattern = re.compile(r'\s[IVXLCDM]{2,}[\s,.?!)"\'\]>]')
+    pattern = re.compile(rf'\s[IVXLCDM]{{2,}}{punctuation}')
     result = string
     while True:
         match = pattern.search(result)
@@ -117,7 +117,7 @@ def num_to_words(text):
 
 def replace_abbreviations(string):
     # abbreviations 1 to 4 characters long. It will get things like A and I, but those are pronounced with their letter
-    pattern = re.compile(r'(^|[\s("\'\[<])([A-Z]{1,4})([\s,.?!)"\'\]>]|$)')
+    pattern = re.compile(rf'(^|[\s("\'\[<])([A-Z]{{1,4}})({punctuation}|$)')
     result = string
     while True:
         match = pattern.search(result)