kromskii2/alltalk_tts
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #190 from ytt246/main

Browse Source 
Edited text cleaning to support some commonly used fullwidth/CJK punctuations
This commit is contained in:
erew123 authored and GitHub committed 2024-04-28 11:12:32 +01:00
2 parents dd52597f49 5aad7bcc45
commit 1e1b7fdd8c
3 files changed
+27 -12
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

script.py
+8 -4
View File
@@ -698,9 +698,11 @@ def output_modifier(string, state):
print(f"[{params['branding']}TTSGen] {voice_description}")
# Replace multiple exclamation marks, question marks, or other punctuation with a single instance
cleaned_part = re.sub(r"([!?.])\1+", r"\1", part)
cleaned_part = re.sub(r"([!?.\u3002\uFF1F\uFF01\uFF0C])\1+", r"\1", part)
# Replace "Chinese ellipsis" with a single dot
cleaned_part = re.sub(r"\u2026{1,2}", ". ", cleaned_part)
# Further clean to remove any other unwanted characters
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]', '', cleaned_part)
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]', '', cleaned_part)
# Remove all newline characters (single or multiple)
cleaned_part = re.sub(r"\n+", " ", cleaned_part)
@@ -725,9 +727,11 @@ def output_modifier(string, state):
# Decode HTML entities first
cleaned_part = html.unescape(cleaned_string)
# Replace multiple instances of certain punctuation marks with a single instance
cleaned_part = re.sub(r"([!?.])\1+", r"\1", cleaned_part)
cleaned_part = re.sub(r"([!?.\u3002\uFF1F\uFF01\uFF0C])\1+", r"\1", cleaned_part)
# Replace "Chinese ellipsis" with a single dot
cleaned_part = re.sub(r"\u2026{1,2}", ". ", cleaned_part)
# Further clean to remove any other unwanted characters
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]', '', cleaned_part)
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]', '', cleaned_part)
# Remove all newline characters (single or multiple)
cleaned_part = re.sub(r"\n+", " ", cleaned_part)
# Process the part and give it a non-character name if being used vai API or standalone.
system/tts_generator/tts_generator.html
+7 -2
View File
@@ -927,10 +927,15 @@
cleanedText = cleanedText.replace(/\\.\\s'/g, ".'");
cleanedText = cleanedText.replace(/(?<!\\w)\['\]+(?!\\w)/g, '');
cleanedText = cleanedText.replace(/"/g, "'");
cleanedText = cleanedText.replace(/[^a-zA-Z0-9\s.,;:!?\-'"$À-ÿ\u0400-\u04FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]/g, '');
cleanedText = cleanedText.replace(/[^a-zA-Z0-9\s.,;:!?\-'"$À-ÿ\u0400-\u04FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u2026\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]/g, '');
cleanedText = cleanedText.replace(/-\\s/g, '; ');
cleanedText = cleanedText.replace(/--/g, '; ');
let sentences = cleanedText.split(/(?<=\[.!?:\])\\s+/);
// Split by Fullwidth/CJK punctuation regardless of if there are spaces behind them
// Please test if it works properly. In my test they it should be
// (?<=[.!?:])\s+ instead of
// (?<=\[.!?:\])\\s+
// so I don't know whats going on here
let sentences = cleanedText.split(/(?<=[\[.!?:\]])\s+|(?<=[\u3002\uFF01\uFF1A\uFF1F])/);
if (!sentences) {
console.error("No sentences found in the input text.");
tts_server.py
+12 -6
View File
@@ -1055,9 +1055,11 @@ async def tts_generate(
voice_description = "\033[36mCharacter (Text-not-inside)\033[0m" if text_not_inside == "character" else "\033[92mNarrator (Text-not-inside)\033[0m"
print(f"[{params['branding']}TTSGen] {voice_description}")
# Replace multiple exclamation marks, question marks, or other punctuation with a single instance
cleaned_part = re.sub(r'([!?.])\1+', r'\1', part)
cleaned_part = re.sub(r'([!?.\u3002\uFF1F\uFF01\uFF0C])\1+', r'\1', part)
# Replace "Chinese ellipsis" with a single dot
cleaned_part = re.sub(r"\u2026{1,2}", ". ", cleaned_part)
# Further clean to remove any other unwanted characters
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]', '', cleaned_part)
cleaned_part = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]', '', cleaned_part)
# Remove all newline characters (single or multiple)
cleaned_part = re.sub(r'\n+', ' ', cleaned_part)
output_file = this_dir / "outputs" / f"{output_file_name}_{uuid.uuid4()}_{int(time.time())}.wav"
@@ -1086,15 +1088,19 @@ async def tts_generate(
output_cache_url = f'http://{params["ip_address"]}:{params["port_number"]}/audiocache/{output_file_name}.wav'
if text_filtering == "html":
cleaned_string = html.unescape(standard_filtering(text_input))
cleaned_string = re.sub(r'([!?.])\1+', r'\1', text_input)
cleaned_string = re.sub(r'([!?.\u3002\uFF1F\uFF01\uFF0C])\1+', r'\1', text_input)
# Replace "Chinese ellipsis" with a single dot
cleaned_string = re.sub(r"\u2026{1,2}", ". ", cleaned_string)
# Further clean to remove any other unwanted characters
cleaned_string = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]', '', cleaned_string)
cleaned_string = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]', '', cleaned_string)
# Remove all newline characters (single or multiple)
cleaned_string = re.sub(r'\n+', ' ', cleaned_string)
elif text_filtering == "standard":
cleaned_string = re.sub(r'([!?.])\1+', r'\1', text_input)
cleaned_string = re.sub(r'([!?.\u3002\uFF1F\uFF01\uFF0C])\1+', r'\1', text_input)
# Replace "Chinese ellipsis" with a single dot
cleaned_string = re.sub(r"\u2026{1,2}", ". ", cleaned_string)
# Further clean to remove any other unwanted characters
cleaned_string = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F]', '', cleaned_string)
cleaned_string = re.sub(r'[^a-zA-Z0-9\s.,;:!?\-\'"$\u0400-\u04FF\u00C0-\u00FF\u0150\u0151\u0170\u0171\u0900-\u097F\u2018\u2019\u201C\u201D\u3001\u3002\u3040-\u309F\u30A0-\u30FF\u4E00-\u9FFF\u3400-\u4DBF\uF900-\uFAFF\u0600-\u06FF\u0750-\u077F\uFB50-\uFDFF\uFE70-\uFEFF\uAC00-\uD7A3\u1100-\u11FF\u3130-\u318F\uFF01\uFF0c\uFF1A\uFF1B\uFF1F]', '', cleaned_string)
# Remove all newline characters (single or multiple)
cleaned_string = re.sub(r'\n+', ' ', cleaned_string)
else: