.env.example

@@ -1,14 +0,0 @@
-# port exposed on the host — default 10800 (piper=10200, whisper=10300, wakeword=10400)
-WYOMING_PORT=10800
-
-# absolute path to the directory containing your voice reference wav
-VOICE_REF_DIR=/path/to/your/voice
-
-# filename of the wav inside that directory
-VOICE_REF_FILE=reference.wav
-
-# volume multiplier applied to output audio (3.0 = louder, 1.0 = raw)
-VOLUME_BOOST=3.0
-
-# torch device: cuda or cpu
-TORCH_DEVICE=cuda

Dockerfile

@@ -1,25 +0,0 @@
-FROM pytorch/pytorch:2.4.0-cuda12.4-cudnn9-runtime
-
-WORKDIR /app
-
-ENV DEBIAN_FRONTEND=noninteractive \
-    PYTHONDONTWRITEBYTECODE=1 \
-    PYTHONUNBUFFERED=1 \
-    HF_HOME=/cache/huggingface \
-    PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
-
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    libsndfile1 \
-    ffmpeg \
-    && rm -rf /var/lib/apt/lists/*
-
-COPY pyproject.toml .
-COPY wyoming_chatterbox/ ./wyoming_chatterbox/
-
-RUN pip install --no-cache-dir . && \
-    pip install --no-cache-dir chatterbox-tts
-
-VOLUME ["/cache", "/voice"]
-
-ENTRYPOINT ["wyoming-chatterbox"]
-CMD ["--uri", "tcp://0.0.0.0:10800", "--voice-ref", "/voice/reference.wav"]

README.md

@@ -2,100 +2,61 @@
 
 [wyoming protocol](https://github.com/rhasspy/wyoming) server for [chatterbox tts](https://github.com/resemble-ai/chatterbox) with voice cloning.
 
-clone any voice with a 10-30 second audio sample. integrates directly with home assistant as a tts provider.
+clone any voice with a 10-30 second audio sample.
 
 ## requirements
 
-- nvidia gpu with 4gb+ vram (3.5gb used at runtime)
+- nvidia gpu with 4gb+ vram
-- cuda 12.x host driver (≥550.54.14)
+- cuda 12.x
-- [nvidia container toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) installed on host
+- python 3.10+
-- docker + docker compose v2
 
-## docker (recommended)
+## install
-
-### 1. configure
 
+from source:
 ```bash
 git clone https://github.com/sudoxreboot/wyoming-chatterbox
 cd wyoming-chatterbox
-cp .env.example .env
-```
-
-edit `.env`:
-
-```env
-WYOMING_PORT=10800          # host port — change if 10800 is taken
-VOICE_REF_DIR=/path/to/dir  # directory containing your reference wav
-VOICE_REF_FILE=reference.wav
-VOLUME_BOOST=3.0
-TORCH_DEVICE=cuda
-```
-
-### 2. build and run
-
-```bash
-docker compose build
-docker compose up -d
-```
-
-first run downloads ~3.5gb of chatterbox model weights into a named docker volume (`chatterbox-cache`). this only happens once.
-
-### 3. check logs
-
-```bash
-docker compose logs -f
-# you should see: "starting server at tcp://0.0.0.0:10800"
-```
-
-### voice reference tips
-
-- 10-30 seconds of clean speech
-- no background music or noise
-- consistent speaking style
-- wav format (any sample rate)
-
----
-
-## install from source (no docker)
-
-```bash
-git clone https://github.com/sudoxreboot/wyoming-chatterbox
-cd wyoming-chatterbox
-python3 -m venv .venv
-source .venv/bin/activate
 pip install .
 ```
 
+## usage
+
 ```bash
-wyoming-chatterbox --uri tcp://0.0.0.0:10800 --voice-ref /path/to/voice.wav
+wyoming-chatterbox --uri tcp://0.0.0.0:10201 --voice-ref /path/to/voice_sample.wav
 ```
 
 ### options
 
 | option | default | description |
 |--------|---------|-------------|
-| `--uri` | required | server uri (e.g., `tcp://0.0.0.0:10800`) |
+| `--uri` | required | server uri (e.g., `tcp://0.0.0.0:10201`) |
 | `--voice-ref` | required | path to voice reference wav (10-30s of speech) |
 | `--volume-boost` | 3.0 | output volume multiplier |
 | `--device` | cuda | torch device (`cuda` or `cpu`) |
 | `--debug` | false | enable debug logging |
 
----
+## voice reference tips
 
-## systemd service (source install)
+for best results:
+- 10-30 seconds of clean speech
+- no background music or noise
+- consistent speaking style
+- wav format (any sample rate)
+
+## systemd service
 
 ```bash
-sudo tee /etc/systemd/system/wyoming-chatterbox.service << EOF
+sudo tee /etc/systemd/system/wyoming-chatterbox.service << 'EOF'
 [Unit]
 Description=Wyoming Chatterbox TTS
 After=network-online.target
 
 [Service]
 Type=simple
-User=$(whoami)
+User=YOUR_USER
 Environment=PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
-ExecStart=$(pwd)/.venv/bin/wyoming-chatterbox \
+ExecStart=/path/to/venv/bin/wyoming-chatterbox \
-  --uri tcp://0.0.0.0:10800 \
+  --uri tcp://0.0.0.0:10201 \
   --voice-ref /path/to/voice_reference.wav \
   --volume-boost 3.0
 Restart=always
@@ -109,33 +70,25 @@ sudo systemctl daemon-reload
 sudo systemctl enable --now wyoming-chatterbox
 ```
 
----
-
 ## home assistant
 
 1. settings → devices & services → add integration
-2. search **wyoming protocol**
+2. search "wyoming protocol"
-3. host: your server ip, port: `10800` (or whatever you set in `.env`)
+3. host: `YOUR_IP`, port: `10201`
-4. select it as your tts provider in the voice assistant pipeline
+4. use in your voice assistant pipeline as tts
-
----
 
 ## gpu memory
 
-chatterbox uses ~3.5gb vram at runtime. if you get oom errors:
+chatterbox uses ~3.5gb vram. if you get oom errors:
 
 ```bash
+# check gpu usage
 nvidia-smi
 
-# docker
+# kill zombie processes
-docker compose restart
-
-# source
 pkill -f wyoming-chatterbox
 ```
 
----
-
 ## license
 
 mit

compose.yaml

@@ -1,28 +0,0 @@
-services:
-  wyoming-chatterbox:
-    build: .
-    image: wyoming-chatterbox:latest
-    container_name: wyoming-chatterbox
-    restart: unless-stopped
-    ports:
-      - "${WYOMING_PORT:-10800}:10800"
-    volumes:
-      - chatterbox-cache:/cache
-      - ${VOICE_REF_DIR}:/voice:ro
-    environment:
-      - PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
-    command: >
-      --uri tcp://0.0.0.0:10800
-      --voice-ref /voice/${VOICE_REF_FILE}
-      --volume-boost ${VOLUME_BOOST:-3.0}
-      --device ${TORCH_DEVICE:-cuda}
-    deploy:
-      resources:
-        reservations:
-          devices:
-            - driver: nvidia
-              count: 1
-              capabilities: [gpu]
-
-volumes:
-  chatterbox-cache: