Init commit

.vscode/settings.json

@@ -0,0 +1,5 @@
+{
+    "python-envs.defaultEnvManager": "ms-python.python:conda",
+    "python-envs.defaultPackageManager": "ms-python.python:conda",
+    "python-envs.pythonProjects": []
+}

README.md

@@ -4,7 +4,7 @@ emoji: 🏢
 colorFrom: red
 colorTo: gray
 sdk: gradio
-sdk_version: 6.5.1
+sdk_version: 5.28.0
 app_file: app.py
 pinned: false
 short_description: Beat detection for music

app.py

@@ -0,0 +1,102 @@
+import os
+import sys
+import uuid
+from pathlib import Path
+from contextlib import contextmanager
+
+import numpy as np
+import torch
+import gradio as gr
+import librosa
+
+from BeatNet.BeatNet import BeatNet
+
+from pyharp.core import ModelCard, build_endpoint
+from pyharp.media.audio import save_audio
+from pyharp import LabelList, AudioLabel, OutputLabel
+from audiotools import AudioSignal
+
+LOUDNESS_DB = -16.
+SAMPLE_RATE = 48_000
+ENCODEC_SAMPLE_RATE = 16_000
+AUDIOSEAL_SAMPLE_RATE = 16_000
+
+model_card = ModelCard(
+    name="BeatNet Beat Detection",
+    description=("Beat detection for audio."),
+    author="Mojtaba Heydari, Frank Cwitkowitz, Zhiyao Duan",
+    tags=["beat detection"]
+)
+
+print("Initializing BeatNet model...")
+estimator = BeatNet(1, mode="offline", inference_model="DBN", plot=[], thread=False)
+
+def load_audio(audio_path):
+    try:
+        wav, sr = librosa.load(audio_path, mono=True)
+        return wav, sr
+    
+    except Exception as e:
+        print(f"Audio preprocessing failed: {e}")
+        raise ValueError(f"Failed to load audio: {str(e)}")
+
+def process_fn(inp_audio):
+    audio_np, sr = load_audio(inp_audio)
+
+    print(f"sr: {sr}, audio shape: {audio_np.shape}")
+    if audio_np.ndim == 1:
+        audio_np = audio_np[None, None, :]
+    else:
+        audio_np = np.transpose(audio_np, (1, 0))[None, ...]
+
+    print(f"formatted audio: {audio_np.shape}")
+
+    output = estimator.process(audio_np)
+
+    output_labels = LabelList()
+
+    for t, b in output:
+        output_labels.labels.append(
+            AudioLabel(
+                t = t,
+                label = f"{b}",
+                description = f"Beat: {b}",
+                color = OutputLabel.rgb_color_to_int(),
+                amplitude = 1.0
+            )
+        )
+
+    return inp_audio, output_labels
+
+
+with gr.Blocks() as app:
+    gr.Markdown("## BeatNet Beat Detection")
+
+    # Inputs
+    input_audio = gr.Audio(
+        label="Input Audio",
+        type="filepath",
+        sources=["upload", "microphone"]
+    )
+
+    # Outputs
+    output_wav = gr.Audio(
+        type="filepath",
+        label="Watermarked Speech"
+    )
+    output_label = gr.JSON(label="Watermark Confidence")
+
+    _ = build_endpoint(
+        model_card=model_card,
+        input_components=[
+            input_audio
+        ],
+        output_components=[
+            output_wav,
+            output_label
+        ],
+        process_fn=process_fn
+    )
+
+if __name__ == '__main__':
+    app.launch(share=True, show_error=True, debug=True)

requirements.txt

@@ -0,0 +1,4 @@
+git+https://github.com/TEAMuP-dev/pyharp.git@v0.3.0
+librosa
+madmom
+BeatNet