Added italian text cleaner

TTS/tts/utils/text/cleaners.py

@@ -11,6 +11,9 @@
 from .english.number_norm import normalize_numbers as en_normalize_numbers
 from .english.time_norm import expand_time_english
 from .french.abbreviations import abbreviations_fr
+from .italian.abbreviations import abbreviations_it
+from .italian.number_norm import normalize_numbers as it_normalize_numbers
+from .italian.time_norm import expand_time_italian
 
 # Regular expression matching whitespace:
 _whitespace_re = re.compile(r"\s+")
@@ -21,6 +24,8 @@ def expand_abbreviations(text, lang="en"):
         _abbreviations = abbreviations_en
     elif lang == "fr":
         _abbreviations = abbreviations_fr
+    elif lang == "it":
+        _abbreviations = abbreviations_it
     for regex, replacement in _abbreviations:
         text = re.sub(regex, replacement, text)
     return text
@@ -70,6 +75,8 @@ def replace_symbols(text, lang="en"):
         text = text.replace("&", " et ")
     elif lang == "pt":
         text = text.replace("&", " e ")
+    elif lang == "it":
+        text = text.replace("&", " e ")
     elif lang == "ca":
         text = text.replace("&", " i ")
         text = text.replace("'", "")
@@ -130,6 +137,21 @@ def phoneme_cleaners(text):
     return text
 
 
+def italian_cleaners(text):
+    """Pipeline for Italian text: time + light number + abbreviations + symbol cleanup."""
+    # optional: normalize common smart punctuation
+    text = text.replace("’", "'").replace("“", '"').replace("”", '"').replace("–", "-").replace("—", "-")
+
+    text = lowercase(text)
+    text = expand_time_italian(text)
+    text = it_normalize_numbers(text)
+    text = expand_abbreviations(text, lang="it")
+    text = replace_symbols(text, lang="it")
+    text = remove_aux_symbols(text)
+    text = collapse_whitespace(text)
+    return text
+
+
 def french_cleaners(text):
     """Pipeline for French text. There is no need to expand numbers, phonemizer already does that"""
     text = expand_abbreviations(text, lang="fr")

TTS/tts/utils/text/italian/abbreviations.py

@@ -0,0 +1,112 @@
+import re
+
+_END = r"(?=\s|$|[,\.;:\)\]\}\!?])"
+
+
+def _c(pattern: str) -> re.Pattern:
+    """Compile a case-insensitive regex."""
+    return re.compile(pattern, re.IGNORECASE)
+
+
+abbreviations_it = [
+    # --- Courtesy / correspondence
+    (_c(r"\bspett\.?\s*(?:le\.?)?\b" + _END), "spettabile"),
+    (_c(r"\begr\.?\b" + _END), "egregio"),
+    (_c(r"\bgent\.?\s*mo\.?\b" + _END), "gentilissimo"),
+    (_c(r"\bgent\.?\s*ma\.?\b" + _END), "gentilissima"),
+    (_c(r"\bill\.?\s*mo\.?\b" + _END), "illustrissimo"),
+    (_c(r"\bill\.?\s*ma\.?\b" + _END), "illustrissima"),
+    (_c(r"\batt\.?\b" + _END), "attenzione"),
+    (_c(r"\bc\.?\s*a\.?\b" + _END), "cortese attenzione"),  # often "alla c.a."
+    # --- People / titles / professions
+    (_c(r"\bsig\.?\b" + _END), "signor"),
+    (_c(r"\bsig\.?\s*ra\.?\b" + _END), "signora"),
+    (_c(r"\bsig\.?\s*na\.?\b" + _END), "signorina"),
+    (_c(r"\bsigg\.?\b" + _END), "signori"),
+    (_c(r"\bdott\.?\b" + _END), "dottore"),
+    (_c(r"\bdott\.?\s*ssa\.?\b" + _END), "dottoressa"),
+    (_c(r"\bprof\.?\b" + _END), "professore"),
+    (_c(r"\bprof\.?\s*ssa\.?\b" + _END), "professoressa"),
+    (_c(r"\bing\.?\b" + _END), "ingegnere"),
+    (_c(r"\barch\.?\b" + _END), "architetto"),
+    (_c(r"\bavv\.?\b" + _END), "avvocato"),
+    (_c(r"\bgeom\.?\b" + _END), "geometra"),
+    (_c(r"\brag\.?\b" + _END), "ragioniere"),
+    (_c(r"\bcomm\.?\b" + _END), "commercialista"),
+    (_c(r"\bdott\.?\s*comm\.?\b" + _END), "dottore commercialista"),
+    # --- Generic shorthand / Latin-ish
+    (_c(r"\becc\.?\b" + _END), "eccetera"),
+    (_c(r"\betc\.?\b" + _END), "eccetera"),
+    (_c(r"\bes\.?\b" + _END), "per esempio"),
+    (_c(r"\bp\.?\s*es\.?\b" + _END), "per esempio"),
+    (_c(r"\bad\.?\s*es\.?\b" + _END), "ad esempio"),
+    (_c(r"\bcioe\.?\b" + _END), "cioè"),
+    (_c(r"\boss\.?\b" + _END), "ossia"),
+    (_c(r"\bcfr\.?\b" + _END), "confronta"),
+    (_c(r"\bvd\.?\b" + _END), "vedi"),
+    (_c(r"\bvv\.?\b" + _END), "vedi"),
+    # --- Document structure / references (useful for academic text)
+    (_c(r"\bfig\.?\b" + _END), "figura"),
+    (_c(r"\btab\.?\b" + _END), "tabella"),
+    (_c(r"\beq\.?\b" + _END), "equazione"),
+    (_c(r"\bsez\.?\b" + _END), "sezione"),
+    (_c(r"\bsec\.?\b" + _END), "sezione"),
+    (_c(r"\bcap\.?\b" + _END), "capitolo"),
+    (_c(r"\bpar\.?\b" + _END), "paragrafo"),
+    (_c(r"\bapp\.?\b" + _END), "appendice"),
+    (_c(r"\bpagg\.?\b" + _END), "pagine"),
+    (_c(r"\bpag\.?\b" + _END), "pagina"),
+    # --- Identifiers / contact / numbering
+    (_c(r"\btel\.?\b" + _END), "telefono"),
+    (_c(r"\bcell\.?\b" + _END), "cellulare"),
+    (_c(r"\bint\.?\b" + _END), "interno"),
+    (_c(r"\bn\.?\b" + _END), "numero"),
+    (_c(r"\bnn\.?\b" + _END), "numeri"),
+    (_c(r"\bnr\.?\b" + _END), "numero"),
+    (_c(r"\bcod\.?\s*fisc\.?\b" + _END), "codice fiscale"),
+    (_c(r"\bc\.?\s*f\.?\b" + _END), "codice fiscale"),
+    (_c(r"\bp\.?\s*iva\.?\b" + _END), "partita iva"),
+    (_c(r"\bpiva\b" + _END), "partita iva"),
+    # --- Dates / formal ref (common in letters)
+    (_c(r"\bu\.?\s*s\.?\b" + _END), "ultimo scorso"),
+    (_c(r"\bp\.?\s*v\.?\b" + _END), "prossimo venturo"),
+    # --- Company legal forms (common in names)
+    (_c(r"\bs\.?\s*p\.?\s*a\.?\b" + _END), "società per azioni"),
+    (_c(r"\bs\.?\s*r\.?\s*l\.?\b" + _END), "società a responsabilità limitata"),
+    (_c(r"\bs\.?\s*n\.?\s*c\.?\b" + _END), "società in nome collettivo"),
+    (_c(r"\bs\.?\s*a\.?\s*s\.?\b" + _END), "società in accomandita semplice"),
+    (_c(r"\bcoop\.?\b" + _END), "cooperativa"),
+    (_c(r"\bonlus\b" + _END), "organizzazione non lucrativa di utilità sociale"),
+    # --- Months (common in dates)
+    (_c(r"\bgen\.?\b" + _END), "gennaio"),
+    (_c(r"\bfeb\.?\b" + _END), "febbraio"),
+    (_c(r"\bmar\.?\b" + _END), "marzo"),
+    (_c(r"\bapr\.?\b" + _END), "aprile"),
+    (_c(r"\bmag\.?\b" + _END), "maggio"),
+    (_c(r"\bgiu\.?\b" + _END), "giugno"),
+    (_c(r"\blug\.?\b" + _END), "luglio"),
+    (_c(r"\bago\.?\b" + _END), "agosto"),
+    (_c(r"\bset\.?\b" + _END), "settembre"),
+    (_c(r"\bott\.?\b" + _END), "ottobre"),
+    (_c(r"\bnov\.?\b" + _END), "novembre"),
+    (_c(r"\bdic\.?\b" + _END), "dicembre"),
+    # --- Web/contact tokens (optional but often helpful in prompts)
+    (_c(r"\be-?mail\b" + _END), "email"),
+    (_c(r"\bwww\.?\b" + _END), "doppia vu doppia vu doppia vu"),
+]
+
+# More aggressive abbreviations
+abbreviations_it_aggressive = [
+    # Address components
+    (_c(r"\bv\.?\s*le\.?\b" + _END), "viale"),  # v.le
+    (_c(r"\bp\.?\s*za\.?\b" + _END), "piazza"),  # p.za
+    (_c(r"\bp\.?\s*zza\.?\b" + _END), "piazza"),  # p.zza
+    (_c(r"\bp\.?\s*le\.?\b" + _END), "piazzale"),  # p.le
+    (_c(r"\bc\.?\s*so\.?\b" + _END), "corso"),  # c.so
+    # Saints: expand "S." to "san" only when followed by a capitalized token
+    # (helps avoid matching "s." in arbitrary contexts).
+    (_c(r"\bs\.(?=\s+[A-ZÀ-ÖØ-Ý])"), "san"),
+    (_c(r"\bs\.\s*maria\b"), "santa maria"),
+]
+
+abbreviations_it_all = abbreviations_it + abbreviations_it_aggressive

TTS/tts/utils/text/italian/number_norm.py

@@ -0,0 +1,32 @@
+import re
+
+# 1.234.567 -> 1234567  (Italian thousands separator)
+_THOUSANDS_DOT_RE = re.compile(r"\b(\d{1,3})(\.\d{3})+\b")
+
+# decimal comma: 3,14 -> 3 virgola 14
+_DECIMAL_COMMA_RE = re.compile(r"(?<=\d),(?=\d)")
+
+# decimal dot (only when it doesn't look like thousands groups): 3.14 -> 3 virgola 14
+_DECIMAL_DOT_RE = re.compile(r"(?<=\d)\.(?=\d)")
+
+
+def normalize_numbers(text: str) -> str:
+    # Remove thousands separators like 1.234.567
+    def _rm_thousands(m: re.Match) -> str:
+        return m.group(0).replace(".", "")
+
+    text = _THOUSANDS_DOT_RE.sub(_rm_thousands, text)
+
+    # Percent / currency / degree signs (simple, robust)
+    text = re.sub(r"(\d)\s*%", r"\1 percento", text)
+    text = re.sub(r"€\s*(\d+(?:[.,]\d+)?)", r"\1 euro", text)
+    text = re.sub(r"(\d+(?:[.,]\d+)?)\s*€", r"\1 euro", text)
+    text = re.sub(r"(\d+(?:[.,]\d+)?)\s*°C\b", r"\1 gradi celsius", text, flags=re.IGNORECASE)
+    text = re.sub(r"(\d+(?:[.,]\d+)?)\s*°\b", r"\1 gradi", text)
+
+    # Convert decimal separators between digits to spoken form.
+    # Order matters: after removing thousands dots, remaining dot between digits is likely decimal.
+    text = _DECIMAL_COMMA_RE.sub(" virgola ", text)
+    text = _DECIMAL_DOT_RE.sub(" virgola ", text)
+
+    return text

TTS/tts/utils/text/italian/time_norm.py

@@ -0,0 +1,63 @@
+import re
+
+_TIME_RE = re.compile(r"\b([01]?\d|2[0-3])[:\.]([0-5]\d)\b")
+
+# 0..59 (sufficient for minutes)
+_IT_0_19 = [
+    "zero",
+    "uno",
+    "due",
+    "tre",
+    "quattro",
+    "cinque",
+    "sei",
+    "sette",
+    "otto",
+    "nove",
+    "dieci",
+    "undici",
+    "dodici",
+    "tredici",
+    "quattordici",
+    "quindici",
+    "sedici",
+    "diciassette",
+    "diciotto",
+    "diciannove",
+]
+_IT_TENS = {
+    20: "venti",
+    30: "trenta",
+    40: "quaranta",
+    50: "cinquanta",
+}
+
+
+def _it_number_0_59(n: int) -> str:
+    if 0 <= n < 20:
+        return _IT_0_19[n]
+    tens = (n // 10) * 10
+    unit = n % 10
+    if tens not in _IT_TENS:
+        return str(n)
+    if unit == 0:
+        return _IT_TENS[tens]
+    # Keep it simple (no elision rules like "ventuno/ventotto"); XTTS usually handles both.
+    return f"{_IT_TENS[tens]} { _IT_0_19[unit] }"
+
+
+def expand_time_italian(text: str) -> str:
+    """
+    09:00 -> 9
+    09:05 -> 9 e cinque
+    14.30 -> 14 e trenta
+    """
+
+    def repl(m: re.Match) -> str:
+        hh = int(m.group(1))
+        mm = int(m.group(2))
+        if mm == 0:
+            return f"{hh}"
+        return f"{hh} e {_it_number_0_59(mm)}"
+
+    return _TIME_RE.sub(repl, text)

tests/text_tests/test_text_cleaners.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 
-from TTS.tts.utils.text.cleaners import english_cleaners, phoneme_cleaners
+from TTS.tts.utils.text.cleaners import english_cleaners, italian_cleaners, phoneme_cleaners
 
 
 def test_time() -> None:
@@ -19,3 +19,15 @@ def test_currency() -> None:
 def test_expand_numbers() -> None:
     assert phoneme_cleaners("-1") == "minus one"
     assert phoneme_cleaners("1") == "one"
+
+
+def test_italian_cleaners_numbers_and_abbreviations() -> None:
+    text = "Alle 09:05 Sig. Bianchi ha pagato €1.234,50 per 50%"
+    expected = "alle 9 e cinque signor bianchi ha pagato 1234 virgola 50 euro per 50 percento"
+    assert italian_cleaners(text) == expected
+
+
+def test_italian_cleaners_temperature_and_time() -> None:
+    text = "Temperatura: 3.5\N{DEGREE SIGN}C alle 14.00"
+    expected = "temperatura, 3 virgola 5 gradi celsius alle 14"
+    assert italian_cleaners(text) == expected