Switch moshi packaging to uv lockfile

Dockerfile

@@ -14,7 +14,6 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
 WORKDIR /app/moshi/
 
 COPY moshi/ /app/moshi/
-RUN uv venv /app/moshi/.venv --python 3.12
 RUN uv sync
 
 RUN mkdir -p /app/ssl

README.md

@@ -17,7 +17,10 @@ PersonaPlex is a real-time, full-duplex speech-to-speech conversational model th
 
 ### Prerequisites
 
-Install the [Opus audio codec](https://github.com/xiph/opus) development library:
+1. Install [uv](https://docs.astral.sh/uv/getting-started/installation/)
+
+2. Install the [Opus audio codec](https://github.com/xiph/opus) development library:
+
 ```bash
 # Ubuntu/Debian
 sudo apt install libopus-dev
@@ -33,12 +36,13 @@ brew install opus
 
 Download this repository and install with:
 ```bash
-pip install moshi/.
+cd moshi
+uv sync
 ```
 
 Extra step for Blackwell based GPUs as suggested in (See https://github.com/NVIDIA/personaplex/issues/2):
 ```bash
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu130
+uv sync --extra cu130
 ```
 
 
@@ -53,12 +57,12 @@ export HF_TOKEN=<YOUR_HUGGINGFACE_TOKEN>
 
 Launch server for live interaction (temporary SSL certs for https):
 ```bash
-SSL_DIR=$(mktemp -d); python -m moshi.server --ssl "$SSL_DIR"
+SSL_DIR=$(mktemp -d); uv run moshi-server --ssl "$SSL_DIR"
 ```
 
 **CPU Offload:** If your GPU has insufficient memory, use the `--cpu-offload` flag to offload model layers to CPU. This requires the `accelerate` package (`pip install accelerate`):
 ```bash
-SSL_DIR=$(mktemp -d); python -m moshi.server --ssl "$SSL_DIR" --cpu-offload
+SSL_DIR=$(mktemp -d); uv run moshi-server --ssl "$SSL_DIR" --cpu-offload
 ```
 
 Access the Web UI from a browser at `localhost:8998` if running locally, otherwise look for the access link printed by the script:
@@ -75,7 +79,7 @@ Add `--cpu-offload` to any command below if your GPU has insufficient memory (re
 **Assistant example:**
 ```bash
 HF_TOKEN=<TOKEN> \
-python -m moshi.offline \
+uv run moshi-offline \
   --voice-prompt "NATF2.pt" \
   --input-wav "assets/test/input_assistant.wav" \
   --seed 42424242 \
@@ -86,7 +90,7 @@ python -m moshi.offline \
 **Service example:**
 ```bash
 HF_TOKEN=<TOKEN> \
-python -m moshi.offline \
+uv run moshi-offline \
   --voice-prompt "NATM1.pt" \
   --text-prompt "$(cat assets/test/prompt_service.txt)" \
   --input-wav "assets/test/input_service.wav" \

moshi/.python-version

@@ -0,0 +1 @@
+3.12

moshi/README.md

@@ -0,0 +1,3 @@
+# moshi-personaplex
+
+Local install for moshi-personaplex.

moshi/pyproject.toml

@@ -1,7 +1,12 @@
 [project]
 name = "moshi-personaplex"
-requires-python = ">= 3.10"
+version = "0.1.0"
 description = "Personaplex finetune of moshi to enable role and voice control."
+readme = "README.md"
+authors = [{name = "Rajarshi Roy", email = "rajarshir@nvidia.com"}]
+maintainers = [{name = "Rajarshi Roy", email = "rajarshir@nvidia.com"}]
+license = {text = "MIT"}
+requires-python = ">= 3.10"
 dependencies = [
     "numpy >= 1.26, < 2.2",
     "safetensors >= 0.4.0, < 0.5",
@@ -10,22 +15,36 @@ dependencies = [
     "sentencepiece == 0.2",
     "sounddevice == 0.5",
     "sphn >= 0.1.4, < 0.2",
-    "torch >= 2.2.0, < 2.5",
+    "torch >= 2.2.0",
     "aiohttp>=3.10.5, <3.11",
 ]
-authors = [{name="Rajarshi Roy", email="rajarshir@nvidia.com"}]
-maintainers = [{name="Rajarshi Roy", email="rajarshir@nvidia.com"}]
-license = {text = "MIT"}
-dynamic = ["version"]
-readme = "README.md"
+
+[build-system]
+requires = ["uv_build>=0.9.13,<0.10.0"]
+build-backend = "uv_build"
+
+[tool.uv.build-backend]
+module-name = "moshi"
+module-root = ""
 
 [project.scripts]
 moshi-server = "moshi.server:main"
 moshi-offline = "moshi.offline:main"
 
-[tool.setuptools.dynamic]
-version = {attr = "moshi.__version__"}
 
-[build-system]
-requires = ["setuptools"]
-build-backend = "setuptools.build_meta"
+# CUDA Config
+
+[[tool.uv.index]]
+name = "pytorch-cu130"
+url = "https://download.pytorch.org/whl/cu130"
+explicit = true
+
+[tool.uv.sources]
+torch = [
+  { index = "pytorch-cu130", extra = "cu130" },
+]
+
+[project.optional-dependencies]
+cu130 = [
+  "torch >= 2.2.0",
+]