Switch to CUDA 12.8 + ONNX-based VAD for RTX 5090 Blackwell support

Upgrade PyTorch to 2.7+ with cu128 wheels for Blackwell (sm_120) GPU
support. Replace silero-vad (which depends on torchaudio) with a direct
ONNX Runtime implementation of the same Silero VAD model, eliminating
the torchaudio dependency entirely.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Dockerfile

@@ -1,4 +1,4 @@
-FROM nvidia/cuda:12.1.1-cudnn8-devel-ubuntu22.04
+FROM nvidia/cuda:12.8.0-cudnn-devel-ubuntu22.04
 
 ENV DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
@@ -12,9 +12,6 @@ RUN apt-get update && apt-get install -y \
     git \
     ffmpeg \
     curl \
-    cmake \
-    ninja-build \
-    build-essential \
     && rm -rf /var/lib/apt/lists/*
 
 # Bootstrap pip for python3.11 (Debian disables ensurepip for system Python)
@@ -24,23 +21,15 @@ RUN ln -sf /usr/bin/python3.11 /usr/bin/python
 
 WORKDIR /app
 
-# Build PyTorch from source with Blackwell (sm_120) support
-RUN git clone --depth 1 https://github.com/pytorch/pytorch.git /tmp/pytorch && \
-    cd /tmp/pytorch && \
-    git submodule update --init --recursive && \
-    TORCH_CUDA_ARCH_LIST="5.0;6.0;7.0;7.5;8.0;8.6;9.0;9.0a;12.0" \
-    python3.11 setup.py install && \
-    cd / && rm -rf /tmp/pytorch
-
-# Install torchvision and torchaudio with CUDA 12.1 support
+# Install PyTorch 2.7+ with CUDA 12.8 support (includes Blackwell/sm_120 support)
 RUN python3.11 -m pip install --no-cache-dir \
-    torchvision torchaudio \
-    --index-url https://download.pytorch.org/whl/cu121
+    torch torchvision \
+    --index-url https://download.pytorch.org/whl/cu128
 
-# Install auto-gptq pre-built wheel for CUDA 12.1 (avoids compiling from source)
+# Install auto-gptq pre-built wheel for CUDA 12.8 (avoids compiling from source)
 RUN python3.11 -m pip install --no-cache-dir \
     "auto-gptq>=0.7.1" \
-    --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu121/
+    --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu128/
 
 # Install the rest of the app requirements
 COPY requirements.txt .