hello

Dockerfile

@@ -0,0 +1,16 @@
+# Read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM python:3.9
+
+RUN useradd -m -u 1000 user
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR /app
+
+COPY --chown=user ./requirements.txt requirements.txt
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . /app
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

_app.py

@@ -0,0 +1,26 @@
+from fastapi import FastAPI, File, UploadFile
+from model_handler import MySeparationModel
+import torchaudio
+
+app = FastAPI()
+model = MySeparationModel()
+
+
+@app.post("/separate/")
+async def separate_audio(file: UploadFile = File(...)):
+    # Save input
+    input_path = "input.wav"
+    with open(input_path, "wb") as f:
+        f.write(await file.read())
+
+    # Process
+    separated = model.predict(input_path)
+
+    # Save outputs
+    output_paths = []
+    for i in range(separated.shape[-1]):
+        path = f"output_{i}.wav"
+        torchaudio.save(path, separated[..., i].detach().cpu(), 8000)
+        output_paths.append(path)
+
+    return {"outputs": output_paths}

app.py

@@ -0,0 +1,8 @@
+from fastapi import FastAPI
+
+app = FastAPI()
+
+
+@app.get("/")
+def greet_json():
+    return {"Hello": "World!"}

auth.py

@@ -0,0 +1,12 @@
+from huggingface_hub import login
+import os
+from dotenv import load_dotenv
+
+load_dotenv()
+
+hf_key = os.getenv('HF_API_KEY')
+
+login(token=hf_key)  # From Profile → Access Tokens
+
+# path = os.environ["PATH"]
+# print(path)

hf_config.yaml

@@ -0,0 +1,4 @@
+# hf_config.yaml
+runtime:
+  dependencies_file: requirements.txt
+  python_version: "3.10"

model_handler.py

@@ -0,0 +1,13 @@
+# model_handler.py
+from speechbrain.inference.separation import SepformerSeparation
+
+
+class MySeparationModel:
+    def __init__(self):
+        self.model = SepformerSeparation.from_hparams(
+            source="speechbrain/sepformer-wham",
+            savedir="pretrained_models/sepformer-wham"
+        )
+
+    def predict(self, audio_path):
+        return self.model.separate_file(audio_path)