Update PaitentVoiceToText.py

PaitentVoiceToText.py

@@ -1,10 +1,9 @@
-# stt_gradio.py
+# PaitentVoiceToText.py
 import torch
 from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
-import gradio as gr
 
 # -------------------
-# 1️⃣ Detect GPU
+# 1️⃣ Detect device
 # -------------------
 use_cuda = torch.cuda.is_available()
 dtype = torch.float16 if use_cuda else torch.float32
@@ -20,7 +19,7 @@ print("⏳ Loading model...")
 model = AutoModelForSpeechSeq2Seq.from_pretrained(
     hub_id,
     torch_dtype=dtype,
-    device_map="auto",          # accelerate handles device placement
+    device_map="auto",         # accelerate manages device placement
     trust_remote_code=True
 )
 
@@ -42,25 +41,14 @@ pipe = pipeline(
 print("🎧 Whisper pipeline ready.")
 
 # -------------------
-# 4️⃣ Transcription Function
+# 4️⃣ Function for external import
 # -------------------
-def transcribe(audio):
-    # Gradio audio input returns a file path
-    if audio is None:
+def record_and_transcribe(audio_file):
+    """
+    Transcribe an audio file (path) or recording.
+    Returns the transcribed text.
+    """
+    if audio_file is None:
         return "No audio provided."
-    result = pipe(audio)
+    result = pipe(audio_file)
     return result["text"]
-
-# -------------------
-# 5️⃣ Gradio Interface
-# -------------------
-demo = gr.Interface(
-    fn=transcribe,
-    inputs=gr.Audio(sources=["microphone", "upload"], type="filepath"),
-    outputs="text",
-    title="🎤 Whisper Speech-to-Text",
-    description="Record or upload audio and get real-time transcription using Whisper."
-)
-
-if __name__ == "__main__":
-    demo.launch()