Update app.py

app.py

@@ -30,7 +30,7 @@ def text2image_latent(text, steps, width, height, images, diversity):
     return(image_paths)
 
 
-def speech_to_text(mic=None, file=None):
+def speech_to_text(mic=None, file=None, state=""):
     if mic is not None:
         audio = mic
     elif file is not None:
@@ -38,7 +38,8 @@ def speech_to_text(mic=None, file=None):
     else:
         return "You must either provide a mic recording or a file"
     transcription = asr(audio)["text"]
-    return transcription
+    state += text + " "
+    return state, state
 
 def zero_shot(image, text_input):
     PIL_image = Image.fromarray(np.uint8(image)).convert('RGB')
@@ -66,7 +67,7 @@ with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column():
             audio_file =[
-            gr.Audio(source="microphone", type="filepath", optional=True),
+            gr.Audio(source="microphone", type="filepath", optional=True), "state", 
             gr.Audio(source="upload", type="filepath", optional=True)]
             text = gr.Textbox(label="Text", placeholder="If you dont want to record or upload your voice you can input text here")
             with gr.Row():
@@ -88,8 +89,8 @@ with gr.Blocks() as demo:
                 zero_shot_clf = gr.Button("Classify Image go brr")
 
         
-    speech_to_text.click(speech_to_text, inputs=audio_file, outputs=text)
+    speech_to_text.click(speech_to_text, inputs=audio_file, outputs=[text,"state"])
     get_image_latent.click(text2image_latent, inputs=[text, steps, width, height, images, diversity], outputs=gallery)
     zero_shot_clf.click(zero_shot, inputs=[gallery, text_input], outputs=label)  
       
-demo.launch()
+demo.launch(live=True)