Update app.py

app.py

@@ -1,8 +1,10 @@
 import gradio as gr
+from transformers import pipeline
 import numpy as np
-from google.cloud import speech_v1
-from google.protobuf import timestamp_pb2
+#from google.cloud import speech_v1
+#from google.protobuf import timestamp_pb2
 
+transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
 
 def transcribe(stream, audio_bytes):
     """Transcribe audio bytes to text using Google Cloud Speech to Text."""
@@ -11,27 +13,10 @@ def transcribe(stream, audio_bytes):
     y = y.astype(np.float32)
     y /= np.max(np.abs(y))
     if stream is not None:
-        # Crea un cliente de Speech to Text
-        client = speech_v1.SpeechClient()
-    
-        # Configura la configuración de la solicitud
-        config = speech_v1.RecognitionConfig()
-        config.language_code = "es-ES"
-        config.encoding = speech_v1.RecognitionConfig.Encoding.LINEAR16
-        config.sample_rate_hertz = 16000
-    
-        # Crea una solicitud de reconocimiento de audio
-        audio = speech_v1.RecognitionAudio(content=audio_bytes)
-        request = speech_v1.RecognizeSpeechRequest(config=config, audio=audio)
-
-        # Realiza la transcripción
-        response = client.recognize_speech(request)
-    
-        # Extrae el texto transcrito
-        transcript = response.results[0].alternatives[0].transcript
+        stream = np.concatenate([stream, y])
     else:
         stream = y
-    return stream, transcript
+    return stream, transcriber({"sampling_rate": sr, "raw": stream})["text"]
 
 
 demo = gr.Interface(