Update app.py

app.py

@@ -1,127 +1,38 @@
-import os, time, requests, gradio as gr
-
-# ---------- Hot-patch to bypass Gradio 4.44.0 JSON-schema bug ----------
-import gradio.blocks as _blocks
-if not hasattr(_blocks.Blocks, "_api_info_patched"):
-    _blocks.Blocks._api_info_patched = True
-    _blocks.Blocks.get_api_info = lambda self: {}
-
-print("Gradio version:", gr.__version__)   # should be 4.44.0
+import gradio as gr
+import os, time, requests
 
 # ---------- Environment Variables ----------
 ASR_API_URL = os.getenv("ASR_API_URL")
 AUTH_TOKEN  = os.getenv("AUTH_TOKEN")
+
 if not ASR_API_URL or not AUTH_TOKEN:
     print("⚠️  ASR_API_URL or AUTH_TOKEN is not set; API calls will fail.")
 
-# ---------- Core Transcription Function ----------
-def transcribe_audio(file_path: str):
-    if not ASR_API_URL or not AUTH_TOKEN:
-        return "❌ Error: ASR_API_URL or AUTH_TOKEN is not set.", ""
-
+def transcribe_audio(audio_file):
     headers = {
-        "accept": "application/json",
         "Authorization": f"Bearer {AUTH_TOKEN}",
     }
-
-    start = time.time()
-    try:
-        with open(file_path, "rb") as f:
-            files = {"file": (os.path.basename(file_path), f, "audio/mpeg")}
-            resp = requests.post(ASR_API_URL, headers=headers, files=files, timeout=120)
-    except Exception as e:
-        return f"❌ Error while calling ASR API: {e}", ""
-
-    elapsed = time.time() - start
-    if resp.status_code == 200:
-        data = resp.json()
-        text = data.get("transcription", "No transcription returned.")
-        return text, f"{data.get('time', elapsed):.2f} s"
-    return f"❌ Error: {resp.status_code}, {resp.text}", ""
-
-# ---------- Styling ----------
-VIOLET_MAIN = "#7F3FBF"   # primary violet
-VIOLET_LIGHT = "#C3A6FF"  # lighter violet for gradient/badge
-
-custom_css = f"""
-#gooya-title {{
-  color:#fff;
-  background:linear-gradient(90deg,{VIOLET_MAIN} 0%,{VIOLET_LIGHT} 100%);
-  border-radius:12px;padding:20px 10px;margin-bottom:12px;
-}}
-.gooya-badge {{
-  display:inline-block;background:{VIOLET_MAIN};color:#fff;
-  border-radius:16px;padding:6px 16px;font-size:.97rem;margin-top:4px;
-}}
-"""
-
-# ---------- UI ----------
-with gr.Blocks(css=custom_css, title="Gooya ASR v1.4") as demo:
-    gr.HTML(
-        f"""
-        <div id="gooya-title">
-          <h1 style='margin-bottom:10px;font-weight:800;font-size:2rem;'>
-              Gooya ASR <span style="font-size:1.1rem;font-weight:400;opacity:.8;">v1.4</span>
-          </h1>
-          <p style='font-size:1.12rem;margin-bottom:2px;'>
-              High-performance Persian Speech-to-Text
-          </p>
-          <p style='font-size:.98rem;color:#e9dbff'>
-              Upload or record a Persian audio file (max&nbsp;30&nbsp;s) and instantly get the transcription.
-          </p>
-        </div>
-        """
-    )
-
-    with gr.Row():
-        with gr.Column():
-            audio_input = gr.Audio(
-                label="Audio Input (upload or record, up to 30 s)",
-                type="filepath",
-                sources=["upload", "microphone"],
-            )
-        with gr.Column():
-            processing_time_tb = gr.Textbox(
-                label="⏱️ Processing Time",
-                interactive=False,
-                elem_classes="gooya-badge",
-            )
-            transcription_tb = gr.Textbox(
-                label="📝 Transcription",
-                lines=5,
-                show_copy_button=True,
-                placeholder="The transcription will appear here...",
-                elem_id="gooya-textbox",
-            )
-
-    with gr.Row():
-        btn_transcribe = gr.Button("Transcribe", variant="primary")
-        btn_clear      = gr.Button("Clear",      variant="secondary")
-
-    gr.Markdown(
-        """
-**Guidelines**  
-- Maximum audio length: **30 seconds**  
-- Audio content should be in Persian.  
-- Both transcription and processing time are displayed immediately.  
-
-See the [Persian ASR Leaderboard](https://huggingface.co/spaces/navidved/open_persian_asr_leaderboard) for benchmarks.
-"""
-    )
-
-    # ---------- Callbacks ----------
-    btn_transcribe.click(
-        transcribe_audio,
-        inputs=audio_input,
-        outputs=[transcription_tb, processing_time_tb],
-    )
-
-    btn_clear.click(
-        lambda: ("", "", None),
-        inputs=None,
-        outputs=[transcription_tb, processing_time_tb, audio_input],
-    )
-
-# ---------- Launch ----------
-if __name__ == "__main__":
-    demo.queue().launch(show_api=True, share=True, debug=True)
+    files = {'file': open(audio_file, 'rb')}
+
+    response = requests.post(ASR_API_URL, headers=headers, files=files)
+    if response.status_code == 200:
+        return response.json().get('transcription', '')
+    else:
+        return f"Error: {response.text}"
+
+with gr.Blocks() as interface:
+    gr.Markdown("# Whisper Large V3 Speech Recognition")
+    gr.Markdown("Upload an audio file or use your microphone to transcribe speech to text.")
+    
+    # Create the input and output components
+    audio_input = gr.Audio(type="filepath", label="Input Audio")
+    output_text = gr.Textbox(label="Transcription")
+    
+    # Add a button to trigger the transcription
+    transcribe_button = gr.Button("Transcribe")
+    
+    # Bind the transcribe_audio function to the button click
+    transcribe_button.click(fn=transcribe_audio, inputs=audio_input, outputs=output_text)
+
+# Launch the Gradio app
+interface.launch(share=True)