Migration towards a python package version.

.gitmodules

@@ -1,3 +0,0 @@
-[submodule "VTC-2"]
-	path = VTC-2
-	url = https://huggingface.co/coml/VTC-2

README.md

@@ -30,8 +30,7 @@ As system dependencies, ensure that [uv](https://docs.astral.sh/uv/), [ffmpeg](h
 You can now clone the repo with:
 
 ```bash
-git lfs install
-git clone --recurse-submodules https://github.com/LAAC-LSCP/VTC.git
+git clone https://github.com/LAAC-LSCP/VTC.git
 cd VTC
 ```
 
@@ -48,7 +47,7 @@ If not, you can use the `scripts/convert.py` file to convert your audios to 16 0
 
 
 ```bash
-uv run scripts/infer.py \
+uv run vtc-infer \
     --wavs audios \        # path to the folder containing the audio files
     --output predictions \ # output folder
     --device cpu           # device to run the model on: ('cpu', 'cuda' or 'gpu', 'mps')
@@ -66,10 +65,10 @@ The model outputs are saved to `<output_folder>/` with the following structure:
 ```
 
 #### Helper script
-An example of a bash script is given to perform inference in `scripts/run.sh`. Simply set the correct variables in the script and run it:
+An example of a bash script is given to perform inference in `run.sh`. Simply set the correct variables in the script and run it:
 
 ```bash
-sh scripts/run.sh
+sh run.sh
 ````
 
 ## 3. Model Performance

check_sys_dependencies.sh

@@ -13,13 +13,6 @@ else
     missing_deps+=("uv")
 fi
 
-# Check for git-lfs
-if command -v git-lfs &> /dev/null; then
-    echo "✓ git-lfs is installed ($(git-lfs --version))"
-else
-    echo "✗ git-lfs is NOT installed"
-    missing_deps+=("git-lfs")
-fi
 
 # Check for ffmpeg
 if command -v ffmpeg &> /dev/null; then

pyproject.toml

@@ -5,18 +5,33 @@ description = "The inference code for the Voice Type Classifier model"
 readme = "README.md"
 requires-python = ">=3.13"
 dependencies = [
+    "huggingface-hub>=0.35.0",
     "pyannote-audio>=3.4.0",
     "segma",
     "tqdm>=4.67.1",
 ]
 
+[project.scripts]
+vtc-infer = "vtc_inference.cli:vtc_infer_cli"
+convert-audios = "vtc_inference.convert_audios:convert_audios_cli"
+
+[project.urls]
+documentation = "https://docs.cognitive-ml.fr/fastabx"
+
 [tool.uv.sources]
 segma = { git = "https://github.com/MarvinLvn/segma", rev = "e6cebc5b232ac1a30554caa7ec35dd22057d07f0" }
 
 [dependency-groups]
+dev = [
+    "ipython>=9.15.0",
+]
 doc = [
     "zensical>=0.0.37",
 ]
 
-[project.urls]
-documentation = "https://docs.cognitive-ml.fr/fastabx"
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/vtc_inference"]

scripts/run.sh → run.sh

@@ -6,6 +6,6 @@ output=out/$(date +%Y%m%d_%H%M)_inference_output
 
 mkdir -p $output
 
-uv run scripts/infer.py \
+uv run vtc-infer \
     --wavs $audios_path \
     --output $output