Update app.py

app.py

@@ -1,12 +1,11 @@
-import streamlit as st
+import gradio as gr
 import os
 from groq import Groq
 import soundfile as sf
 from tempfile import NamedTemporaryFile
-import io  # Für Bytes-IO hinzugefügt
 
 # Load the API key from the environment variable
-api_key = os.getenv('groqwhisper')
+api_key = os.getenv('groq_whisper')
 
 if api_key is None:
     raise ValueError("groq_whisper environment variable is not set")
@@ -14,6 +13,7 @@ if api_key is None:
 # Initialize the Groq client
 client = Groq(api_key=api_key)
 
+
 def processaudio(audio_data):
     try:
         # Entpacken der Audiodaten (Sample-Rate und Numpy-Array)
@@ -39,27 +39,33 @@ def processaudio(audio_data):
     except Exception as e:
         return f"Ein Fehler ist aufgetreten: {str(e)}"
 
-# Streamlit Interface
-st.title("Audio Transkription")
-sr_outputs = st.empty()
-
-# Dateiupload und Mikrofonaufnahme als getrennte Eingaben
-uploaded_file = st.file_uploader("Laden Sie eine Audiodatei hoch", type=["wav", "mp3"])
-audio_bytes = st.audio_input("Oder sprechen Sie jetzt:", type="wav")
 
-if uploaded_file:
-    # Verarbeitung hochgeladener Datei
-    audio_data = sf.read(uploaded_file)
-    transcription = processaudio(audio_data)
-    sr_outputs.text(transcription)
 
-elif audio_bytes:
-    # Verarbeitung Mikrofonaufnahme
+def process_audio(file_path):
+   
     try:
-        # Konvertiere Bytes in Audio-Daten
-        audio_io = io.BytesIO(audio_bytes)
-        audio_data = sf.read(audio_io)
-        transcription = processaudio(audio_data)
-        sr_outputs.text(transcription)
+        # Open the audio file
+        with open(file_path, "rb") as file:
+            # Create a transcription of the audio file
+            transcription = client.audio.transcriptions.create(
+                file=(os.path.basename(file_path), file.read()),  # Correct passing of filename
+                model="whisper-large-v3-turbo",  # Required model to use for transcription
+                prompt="transcribe",  # Optional
+                language="de",  # Optional
+                response_format="json",  # Optional
+                temperature=0.0  # Optional
+            )
+            # Return the transcription text
+            sr_inputs.clear()
+            return transcription.text
     except Exception as e:
-        sr_outputs.text(f"Fehler bei der Aufnahmeverarbeitung: {str(e)}")
+        return f"An error occurred: {str(e)}"
+
+with gr.Blocks() as speech:  
+    with gr.Row():
+        sr_outputs = gr.Textbox(label="Transkription")
+    with gr.Row():
+        sr_inputs = gr.Microphone(type="numpy") 
+    sr_inputs.change(processaudio, inputs=sr_inputs, outputs=sr_outputs)
+
+speech.launch()