Create app.py

app.py

@@ -0,0 +1,55 @@
+# Imports
+import gradio as gr
+import spaces
+import torch
+
+from transformers import pipeline
+
+# Pre-Initialize
+DEVICE = "auto"
+if DEVICE == "auto":
+    DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"[SYSTEM] | Using {DEVICE} type compute device.")
+
+# Variables
+DEFAULT_TASK = "transcribe"
+BATCH_SIZE = 8
+
+repo = pipeline(task="automatic-speech-recognition", model="openai/whisper-large-v3-turbo", chunk_length_s=30, device=DEVICE)
+
+css = '''
+.gradio-container{max-width: 560px !important}
+h1{text-align:center}
+footer {
+    visibility: hidden
+}
+'''
+
+@spaces.GPU(duration=10)
+def transcribe(input=None, task=DEFAULT_TASK):
+    print(input)
+    if input is None: raise gr.Error("Invalid input.")
+    output = repo(input, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
+    return output
+
+def cloud():
+    print("[CLOUD] | Space maintained.")
+
+# Initialize
+with gr.Blocks(css=css) as main:
+    with gr.Column():
+        gr.Markdown("🪄 Transcribe audio to text.")
+        
+    with gr.Column():
+        input = gr.Audio(sources="upload", type="filepath", label="Input")
+        task = gr.Radio(["transcribe", "translate"], label="Task", value=DEFAULT_TASK)
+        submit = gr.Button("▶")
+        maintain = gr.Button("☁️")
+
+    with gr.Column():
+        output = gr.Textbox(lines=1, value="", label="Output")
+            
+    submit.click(transcribe, inputs=[input, task], outputs=[output], queue=False)
+    maintain.click(cloud, inputs=[], outputs=[], queue=False)
+
+main.launch(show_api=True)