Initial commit: OrpheusTail TTS service

- FastAPI service replacing VoiceTail (Bark)
- Emotion tags: <laugh>, <sigh>, <gasp>, etc.
- Voice cloning endpoint (implementation pending)
- Streaming support for head playback
- Same port 8766 for drop-in replacement

Created by Vixy on Day 71 🦊

Dockerfile

@@ -0,0 +1,54 @@
+# OrpheusTail - Orpheus TTS Service for NVIDIA Jetson AGX Orin
+# 
+# Replaces VoiceTail (Bark) with Orpheus for better emotion control
+# and voice cloning capabilities.
+#
+# Based on NVIDIA L4T PyTorch container optimized for Jetson
+
+FROM dustynv/pytorch:2.1-r36.2.0
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    git \
+    wget \
+    libsndfile1 \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY requirements.txt /app/
+
+# Install Python dependencies
+# Note: torch and torchvision are already in the base image
+RUN pip3 install --no-cache-dir -r requirements.txt
+
+# Install orpheus-speech (uses vllm under the hood)
+# Note: vllm version compatibility may need adjustment
+RUN pip3 install orpheus-speech
+
+# Copy application code
+COPY main.py /app/
+
+# Create directories for cache, output, and custom voices
+RUN mkdir -p /app/cache /app/output /app/voices
+
+# Expose API port (same as VoiceTail for drop-in replacement)
+EXPOSE 8766
+
+# Set environment variables
+ENV PYTHONUNBUFFERED=1
+ENV CACHE_DIR=/app/cache
+ENV OUTPUT_DIR=/app/output
+ENV VOICES_DIR=/app/voices
+ENV ORPHEUS_MODEL=canopylabs/orpheus-tts-0.1-finetune-prod
+ENV DEFAULT_VOICE=tara
+ENV MAX_MODEL_LEN=2048
+
+# Health check (longer start period - model loading takes time)
+HEALTHCHECK --interval=30s --timeout=10s --start-period=180s --retries=3 \
+    CMD curl -f http://localhost:8766/health || exit 1
+
+# Run the FastAPI application
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8766"]