fix: Update PyTorch/torchaudio to 2.1.0 for SpeechBrain 1.0 diarization compatibility

- Update torch from 2.0.1 to 2.1.0
- Update torchaudio from 2.0.2 to 2.1.0 (fixes 'NoneType' object has no attribute 'eval' error)
- Update torchvision from 0.15.2 to 0.16.0
- Update pytorch-lightning from 2.0.6 to 2.1.0
- Add explicit speechbrain==1.0.0 dependency
- Update transformers and tokenizers for compatibility
- Update protobuf version constraint

Fixes diarization error: speechbrain.pretrained was deprecated and redirected to speechbrain.inference in SpeechBrain 1.0, but required torchaudio >= 2.1.0

Dockerfile

@@ -20,7 +20,7 @@ RUN pip install --no-cache-dir -r requirements.txt
 
 # Optional: Install CUDA-specific PyTorch if GPU support needed
 # Uncomment and modify for your CUDA version:
-# RUN pip install --force-reinstall torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
+# RUN pip install --force-reinstall torch==2.1.0+cu118 torchvision==0.16.0+cu118 torchaudio==2.1.0+cu118 --index-url https://download.pytorch.org/whl/cu118
 
 # Copy application code
 COPY . .

Dockerfile.gpu

@@ -19,10 +19,11 @@ COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Install CUDA-optimized PyTorch (overwrites CPU versions)
+# Updated to torch 2.1.0+ for SpeechBrain 1.0 / pyannote diarization compatibility
 RUN pip install --force-reinstall \
-    torch==2.0.1+cu118 \
-    torchvision==0.15.2+cu118 \
-    torchaudio==2.0.2+cu118 \
+    torch==2.1.0+cu118 \
+    torchvision==0.16.0+cu118 \
+    torchaudio==2.1.0+cu118 \
     --index-url https://download.pytorch.org/whl/cu118
 
 # Copy application code

requirements.txt

@@ -6,14 +6,15 @@ openai-whisper==20231117
 requests>=2.28.0
 humanize>=4.6.0
 
-# PyTorch ecosystem - pinned for compatibility
-torch==2.0.1
-torchaudio==2.0.2
-torchvision==0.15.2
+# PyTorch ecosystem - updated for SpeechBrain 1.0 compatibility
+# torchaudio >= 2.1.0 is REQUIRED for diarization to work properly
+torch==2.1.0
+torchaudio==2.1.0
+torchvision==0.16.0
 
 # Transformers ecosystem - compatible versions
-transformers==4.30.2
-tokenizers==0.13.3
+transformers==4.35.0
+tokenizers>=0.14.0
 
 # ML dependencies with compatible versions
 numpy==1.24.3
@@ -21,12 +22,14 @@ scipy==1.10.1
 scikit-learn==1.3.0
 
 # Audio processing and ML models
+# Pin speechbrain for stability with pyannote.audio
+speechbrain==1.0.0
 pyannote.audio==3.1.1
-pytorch-lightning==2.0.6
+pytorch-lightning==2.1.0
 
 # Other dependencies
 iso639>=0.1.4
-protobuf>=3.20.0,<4.0.0
+protobuf>=3.20.0,<5.0.0
 matplotlib>=3.5.0
 soundfile>=0.10.3
 ffmpeg-python>=0.2.0