make cudnn optional, suppress transformers 4.48+ warning, update transformers to 4.49.0

Author: bigsk1

README.md

@@ -80,7 +80,7 @@ https://github.com/user-attachments/assets/5581bd53-422b-4a92-9b97-7ee4ea37e09b
     Make sure you have ffmpeg downloaded, on windows terminal ( winget install ffmpeg ) or checkout https://ffmpeg.org/download.html then restart shell or vscode, type ffmpeg -version to see if installed correctly
 
     Local XTTS you also might need cuDNN for using nvidia GPU https://developer.nvidia.com/cudnn  and make sure C:\Program Files\NVIDIA\CUDNN\v9.5\bin\12.6
-is in system PATH or whatever version you downloaded
+is in system PATH or whatever version you downloaded, you can also disable cudnn in the XTTS-V2/config.json to "cudnn_enable": false, if you don't want to use it. 
 
 ### Optional - Download Checkpoints - ONLY IF YOU ARE USING THE LOCAL XTTS
 
@@ -428,6 +428,10 @@ Invalid handle. Cannot load symbol cudnnCreateTensorDescriptor
 ```
 To resolve this:
 
+You can disable cudnn in the XTTS-V2/config.json and set to "cudnn_enable": false,
+
+or
+
 Install cuDNN: Download cuDNN from the NVIDIA cuDNN page https://developer.nvidia.com/cudnn
 
 Here’s how to add it to the PATH:

app/app.py

@@ -22,6 +22,8 @@
 from pydub import AudioSegment
 from .shared import clients
 
+import logging
+logging.getLogger("transformers").setLevel(logging.ERROR)  # transformers 4.48+ warning
 
 # Load environment variables
 load_dotenv()
@@ -60,6 +62,9 @@
 # Check for CUDA availability
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
+# Disable CuDNN explicitly - enable this if you get cudnn errors or change in xtts-v2/config.json
+# torch.backends.cudnn.enabled = False
+
 # Default model size (adjust as needed)
 model_size = "medium.en"
 
@@ -221,8 +226,6 @@ def sync_play_audio(file_path):
 
     pass
 
-# Model and device setup
-# device = 'cuda' if torch.cuda.is_available() else 'cpu'
 output_dir = os.path.join(project_dir, 'outputs')
 os.makedirs(output_dir, exist_ok=True)
 

requirements.txt

@@ -13,7 +13,7 @@ soundfile==0.12.1
 langid==1.1.6
 librosa==0.10.0
 scipy==1.11.4
-transformers==4.41.2
+transformers==4.49.0
 pydantic==2.7.4
 pillow==10.3.0
 

requirements_no_xtts.txt

@@ -11,7 +11,7 @@ soundfile==0.12.1
 langid==1.1.6
 librosa==0.10.0
 scipy==1.11.4
-transformers==4.41.2
+transformers==4.49.0
 pydantic==2.7.4
 pillow==10.3.0