Add TTS/STT service with Coqui TTS and faster-whisper

- Added tts-stt service definition to docker-compose.yml
- Created tts-stt directory with:
  - Dockerfile: Based on debian:trixie-slim with ROCm dependencies
  - requirements.txt: Python packages including TTS, faster-whisper, FastAPI
  - app.py: FastAPI service with /tts and /stt endpoints
- Service includes GPU device mapping (/dev/kfd, /dev/dri) for ROCm acceleration
- Uses YourTTS multilingual model for TTS and large-v3 for Whisper
- Configured to use persistent storage for models and cache

ai/tts-stt/Dockerfile

@@ -0,0 +1,50 @@
+FROM debian:trixie-slim
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    python3-pip \
+    python3-dev \
+    git \
+    wget \
+    ffmpeg \
+    libsndfile1 \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install ROCm runtime libraries (matching host setup)
+RUN apt-get update && apt-get install -y \
+    rocm-Devkits=6.3.0 \
+    rocm-dev=6.3.0 \
+    && rm -rf /var/lib/apt/lists/*
+
+# Create non-root user
+RUN useradd -m -u 1000 appuser
+WORKDIR /app
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Download models (will be cached in volume)
+RUN mkdir -p /app/models && chown appuser:appuser /app/models
+USER appuser
+
+# Install TTS models (Coqui TTS with multilingual model)
+RUN pip install --no-cache-dir TTS && \
+    tts --text "Initializing model..." --model_name tts_models/multilingual/multi-dataset/your_tts --out_path /tmp/init.wav || true
+
+# Install faster-whisper
+RUN pip install --no-cache-dir faster-whisper
+
+# Expose port
+EXPOSE 8000
+
+# Switch to non-root user
+USER appuser
+
+# Copy application code
+COPY app.py .
+
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+    CMD curl -f http://localhost:8000/health || exit 1
+
+# Command to run the application
+CMD ["python", "app.py"]