Update app.py

app.py

@@ -6,6 +6,7 @@ from gradio_client import Client
 from chromadb.config import DEFAULT_DATABASE, DEFAULT_TENANT
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 import os
+import speech_recognition as sr
 os.system("pip install --upgrade gradio")
 
 # Initialisiere ChromaDB
@@ -23,6 +24,22 @@ custom_css = """
 # Verwende die integrierten Embeddings von ChromaDB
 embedding_function = embedding_functions.DefaultEmbeddingFunction()
 
+# Function to transcribe audio data to text
+def transcribe_audio(audio):
+    recognizer = sr.Recognizer()
+    with sr.AudioFile(audio) as source:
+        audio_data = recognizer.record(source)
+        try:
+            text = recognizer.recognize_google(audio_data)
+            return text
+        except sr.UnknownValueError:
+            return "Speech recognition could not understand the audio."
+        except sr.RequestError as e:
+            return f"Could not request results from Google Speech Recognition service; {e}"
+
+
+
+
 client = Client("Qwen/Qwen2.5-72B-Instruct")
 def ask_llm(llm_prompt_input):
     # Erstelle Embedding für den Prompt
@@ -139,7 +156,16 @@ with gr.Blocks() as suche:
     with gr.Row():
         search_button = gr.Button("Suchen")
     search_button.click(search_similar_documents, inputs=prompt_input, outputs=search_output)
-        
+
+
+ with gr.Blocks() as speech:
+    gr.Markdown("### audio", elem_classes="tab-header")
+    with gr.Row():
+        sr_inputs=gr.Audio(source="microphone", type="filepath"),
+        sr_outputs=gr.Textbox(label="Transcribed Text")
+    with gr.Row():
+        submit_button = gr.Button("rec")
+    submit_button.click(transcribe_audio, inputs=sr_inputs, outputs=sr_outputs)   
 
 # Erstelle die Gradio-Schnittstelle
 with gr.Blocks() as demo: