Use speechsynthesizer to play audio in client side.

app.py

@@ -7,13 +7,13 @@ import os
 
 
 def run():
-    os.environ['TOKENIZERS_PARALLELISM'] = 'false'
     st.set_page_config(
         page_title='Visual Question Answering - Bot',
         page_icon=':robot:',
         layout='wide'
     )
 
+    os.environ['TOKENIZERS_PARALLELISM'] = 'false'
     st.session_state['predictor'] = predictor.Predictor()
 
     st.sidebar.title('VQA Bot')

audiobot.py

@@ -4,7 +4,6 @@ from PIL import Image
 from bokeh.models.widgets import Button
 from bokeh.models import CustomJS
 from streamlit_bokeh_events import streamlit_bokeh_events
-import pyttsx3
 
 def show():
     st.title('Visual Question Answering - Audiobot')
@@ -24,7 +23,8 @@ def show():
 
     # Speech recognition based in streamlit based on
     # https://discuss.streamlit.io/t/speech-to-text-on-client-side-using-html5-and-streamlit-bokeh-events/7888
-    stt_button = Button(label='Ask', width=100)
+    stt_button = Button(label='Ask Question', width=100)
+
     stt_button.js_on_event('button_click', CustomJS(code='''
         var recognition = new webkitSpeechRecognition();
         recognition.continuous = false;
@@ -47,7 +47,7 @@ def show():
     result = streamlit_bokeh_events(
         stt_button,
         events='GET_TEXT',
-        key='listen',
+        key='stt_listen',
         refresh_on_update=False,
         override_height=75,
         debounce_time=0)
@@ -56,4 +56,13 @@ def show():
         if 'GET_TEXT' in result:
             answer = st.session_state.predictor.predict_answer_from_text(
                 st.session_state.image, result.get('GET_TEXT'))
-            pyttsx3.speak(answer)
+
+            tts_button = Button(label="Get Answer", width=100)
+            tts_button.js_on_event("button_click", CustomJS(code=f"""
+                var u = new SpeechSynthesisUtterance();
+                u.text = "{answer}";
+                u.lang = 'en-US';
+
+                speechSynthesis.speak(u);
+                """))
+            st.bokeh_chart(tts_button)

packages.txt

@@ -1,2 +0,0 @@
-alsa-utils
-espeak

requirements.txt

@@ -10,7 +10,7 @@ backcall==0.2.0
 beautifulsoup4==4.11.1
 bleach==5.0.0
 blinker==1.4
-bokeh==2.4.3
+bokeh==2.4.1
 cachetools==5.2.0
 certifi==2022.6.15
 cffi==1.15.0
@@ -75,7 +75,6 @@ Pympler==1.0.1
 pyparsing==3.0.9
 pyrsistent==0.18.1
 python-dateutil==2.8.2
-pyttsx3==2.90
 pytz==2022.1
 pytz-deprecation-shim==0.1.0.post0
 PyYAML==6.0