Add Whisper-Large-V3-Turbo Demo

app.py

@@ -1,14 +1,52 @@
-import gradio as gr
 import spaces
 import torch
 
-zero = torch.Tensor([0]).cuda()
-print(zero.device) # <-- 'cpu' 🤔
+import gradio as gr
+from transformers import pipeline
+
+
+MODEL_NAME = "openai/whisper-large-v3-turbo"
+BATCH_SIZE = 8
+FILE_LIMIT_MB = 1000
+
+device = 0 if torch.cuda.is_available() else "cpu"
+
+pipe = pipeline(
+    task="automatic-speech-recognition",
+    model=MODEL_NAME,
+    chunk_length_s=30,
+    device=device,
+)
+
 
 @spaces.GPU
-def greet(n):
-    print(zero.device) # <-- 'cuda:0' 🤗
-    return f"Hello {zero + n} Tensor"
+def transcribe(inputs, task):
+    if inputs is None:
+        raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
+
+    text = pipe(inputs, batch_size=BATCH_SIZE, generate_kwargs={"task": task}, return_timestamps=True)["text"]
+    return  text
+
+
+demo = gr.Blocks(theme=gr.themes.Ocean())
+
+
+with demo:
+    gr.Interface(
+        fn=transcribe,
+        inputs=[
+            gr.Audio(sources="upload", type="filepath", label="Audio file"),
+            gr.Radio(["transcribe", "translate"], label="Task", value="transcribe"),
+        ],
+        outputs="text",
+        title="Whisper Large V3: Transcribe Audio",
+        description=(
+            "Transcribe long-form microphone or audio inputs with the click of a button! Demo uses the"
+            f" checkpoint [{MODEL_NAME}](https://huggingface.co/{MODEL_NAME}) and 🤗 Transformers to transcribe audio files"
+            " of arbitrary length."
+        ),
+        allow_flagging="never",
+    )
+
 
-demo = gr.Interface(fn=greet, inputs=gr.Number(), outputs=gr.Text())
-demo.launch()
+demo.queue().launch(ssr_mode=False)

packages.txt

@@ -0,0 +1 @@
+ffmpeg

requirements.txt

@@ -0,0 +1,2 @@
+transformers
+