Revert "docs: add troubleshooting section for libcudnn dependencies in README"

This reverts commit 6fe0a8784a.

Revert the commit now that the issue is fixed.

Signed-off-by: CHEN, CHUN <jim60105@gmail.com>

README.md

@@ -97,25 +97,6 @@ uv sync --all-extras --dev
 
 You may also need to install ffmpeg, rust etc. Follow openAI instructions here https://github.com/openai/whisper#setup.
 
-### Common Issues & Troubleshooting 🔧
-
-#### libcudnn Dependencies (GPU Users)
-
-If you're using WhisperX with GPU support and encounter errors like:
-
-- `Could not load library libcudnn_ops_infer.so.8`
-- `Unable to load any of {libcudnn_cnn.so.9.1.0, libcudnn_cnn.so.9.1, libcudnn_cnn.so.9, libcudnn_cnn.so}`
-- `libcudnn_ops_infer.so.8: cannot open shared object file: No such file or directory`
-
-This means your system is missing the CUDA Deep Neural Network library (cuDNN). This library is needed for GPU acceleration but isn't always installed by default.
-
-**Install cuDNN (example for apt based systems):**
-
-```bash
-sudo apt update
-sudo apt install libcudnn8 libcudnn8-dev -y
-```
-
 ### Speaker Diarization
 
 To **enable Speaker Diarization**, include your Hugging Face access token (read) that you can generate from [Here](https://huggingface.co/settings/tokens) after the `--hf_token` argument and accept the user agreement for the following models: [Segmentation](https://huggingface.co/pyannote/segmentation-3.0) and [Speaker-Diarization-3.1](https://huggingface.co/pyannote/speaker-diarization-3.1) (if you choose to use Speaker-Diarization 2.x, follow requirements [here](https://huggingface.co/pyannote/speaker-diarization) instead.)
@@ -189,7 +170,7 @@ result = model.transcribe(audio, batch_size=batch_size)
 print(result["segments"]) # before alignment
 
 # delete model if low on GPU resources
-# import gc; gc.collect(); torch.cuda.empty_cache(); del model
+# import gc; import torch; gc.collect(); torch.cuda.empty_cache(); del model
 
 # 2. Align whisper output
 model_a, metadata = whisperx.load_align_model(language_code=result["language"], device=device)
@@ -198,7 +179,7 @@ result = whisperx.align(result["segments"], model_a, metadata, audio, device, re
 print(result["segments"]) # after alignment
 
 # delete model if low on GPU resources
-# import gc; gc.collect(); torch.cuda.empty_cache(); del model_a
+# import gc; import torch; gc.collect(); torch.cuda.empty_cache(); del model_a
 
 # 3. Assign speaker labels
 diarize_model = whisperx.diarize.DiarizationPipeline(use_auth_token=YOUR_HF_TOKEN, device=device)

pyproject.toml

@@ -13,11 +13,11 @@ dependencies = [
     "faster-whisper>=1.1.1",
     "nltk>=3.9.1",
     "numpy>=2.0.2",
-    "onnxruntime>=1.19",
+    "onnxruntime>=1.19,<1.20.0",
     "pandas>=2.2.3",
     "pyannote-audio>=3.3.2",
-    "torch>=2.5.1",
+    "torch<2.4.0",
-    "torchaudio>=2.5.1",
+    "torchaudio",
     "transformers>=4.48.0",
 ]