speech to text fix

app.py

@@ -7,6 +7,7 @@ from transformers import pipeline
 
 asr = pipeline("automatic-speech-recognition")
 latent = gr.Interface.load("spaces/multimodalart/latentdiffusion")
+
 #zero = pipeline("zero-shot-image-classification")
 #zero = gr.Interface.load("spaces/Datatrooper/zero-shot-image-classification")
 #tts = gr.Interface.load("spaces/osanseviero/tortoisse-tts")
@@ -30,7 +31,7 @@ def text2image_latent(text, steps, width, height, images, diversity):
     return(image_paths)
 
 
-def speech_to_text(mic=None, file=None, state=""):
+def speech_to_text(mic=None, file=None):
     if mic is not None:
         audio = mic
     elif file is not None:
@@ -38,7 +39,7 @@ def speech_to_text(mic=None, file=None, state=""):
     else:
         return "You must either provide a mic recording or a file"
     transcription = asr(audio)["text"]
-    return state
+    return transcription
 
     
 with gr.Blocks() as demo: