push some updates to multilingual branch

.gitmodules

@@ -4,6 +4,6 @@
 [submodule "modules/dlas"]
 	path = modules/dlas
 	url = https://github.com/JarodMica/DL-Art-School.git
-[submodule "modules/tortoise-dataset-tools"]
-	path = modules/tortoise-dataset-tools
-	url = https://github.com/JarodMica/tortoise-dataset-tools.git
+[submodule "modules/tortoise_dataset_tools"]
+	path = modules/tortoise_dataset_tools
+	url = https://github.com/JarodMica/tortoise_dataset_tools.git

changelog.md

@@ -10,6 +10,12 @@ AR Quantization
 - TortoiseCPP https://github.com/balisujohn/tortoise.cpp
 
 
+## 3/17/2024
+- Adding in other language capability training to the repo, a few files are modifed in dlas and tortoise-tts modules for the cleaners to allow this to happen.
+- In both DLAS and Tortoise-tts, update the preprocessing text to be able to handle other languages
+- Added https://github.com/JarodMica/tortoise-dataset-tools inside of modules (will be used for other languages)
+    - In this case, made whisperx a mandatory install
+    - Adding a new tab called right now "Prepare Other Langauge" that will allows you to use the dataset tools I used for other languages
 
 ## 2/11/2024
 - Allow for decimal values in the epoch text box as a bandaid to the async gradio issue, causing the training run to crash.  Not sure if it's gradio or if it's an error in the DLAS trainer, but this will need to be fixed as it's quite annoying to have to restart training over and over

modules/tortoise-dataset-tools

@@ -1 +0,0 @@
-Subproject commit eab958e0d844c796279f49b806b788b54d0f3219

modules/tortoise_dataset_tools

@@ -0,0 +1 @@
+Subproject commit 1839a6562c2abb0b9c13f277a98d1f64f50cc050

requirements.txt

@@ -18,3 +18,4 @@ pykakasi
 rotary-embedding-torch==0.4.0
 transformers==4.19
 omegaconf<=2.1
+pysrt

src/train_jard.py

@@ -4,6 +4,11 @@ import argparse
 import yaml
 import datetime
 
+'''
+Note:
+This is a specific file for testing, nothing more than that for Jarod.
+'''
+
 if os.path.exists("runtime"):
 	# Get the directory where the script is located
 	script_dir = os.path.dirname(os.path.abspath(__file__))

src/utils.py

@@ -3185,6 +3185,10 @@ def get_voice_list(dir=get_voice_dir(), append_defaults=False, extensions=["wav"
 	
 	return res
 
+
+    
+    
+
 def get_valle_models(dir="./training/"):
 	return [ f'{dir}/{d}/config.yaml' for d in os.listdir(dir) if os.path.exists(f'{dir}/{d}/config.yaml') ]