add application file

app.py

@@ -0,0 +1,73 @@
+import torch
+from transformers import pipeline
+from transformers.pipelines.audio_utils import ffmpeg_read
+import gradio as gr
+
+MODEL_NAME = "JackismyShephard/whisper-medium.en-finetuned-gtzan"
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="audio-classification",
+    model=MODEL_NAME,
+    device=device,
+)
+
+def classify_audio(filepath):
+    preds = pipe(filepath)
+    outputs = {}
+    for p in preds:
+        outputs[p["label"]] = p["score"]
+    return outputs
+
+
+
+demo = gr.Blocks()
+
+file_transcribe = gr.Interface(
+    fn=transcribe,
+    #TODO not sure we need list here
+    inputs=[
+        #TODO not sure we need '.inputs.'
+        gr.inputs.Audio(source="upload", optional=True, label="Audio file", type="filepath"),
+        #TODO add inputs source upload here, if possible?
+        #TODO add inputs source youtube here, if possible?
+    ],
+    outputs="label", #TODO not sure about this
+    layout="horizontal", #TODO not sure we need this
+    theme="huggingface",
+    title="Classify Genre of Music",
+    description=(
+        "Classify long-form audio or microphone inputs with the click of a button! Demo uses the"
+        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to classify audio files"
+        " of arbitrary length."
+    ),
+    examples=[
+        ["./example.flac", "transcribe", False],
+        ["./example.flac", "transcribe", True],
+    ],
+    cache_examples=True,
+    allow_flagging="never",
+)
+
+mic_transcribe = gr.Interface(
+    fn=transcribe,
+    inputs=[
+        gr.inputs.Audio(source="microphone", type="filepath", optional=True),
+    ],
+    outputs="label", #TODO not sure about this
+    layout="horizontal",
+    theme="huggingface",
+    title="Classify Genre of Music",
+    description=(
+        "Classify long-form audio or microphone inputs with the click of a button! Demo uses the"
+        f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to classify audio files"
+        " of arbitrary length."
+    ),
+    allow_flagging="never",
+)
+
+with demo:
+    gr.TabbedInterface([file_transcribe, mic_transcribe], ["Classify Audio File", "classify Microphone input"])
+
+demo.launch(enable_queue=True)