Update app.py

app.py

@@ -1,49 +1,52 @@
 import streamlit as st
+from gtts import gTTS
+import os
+import base64
 
-# Text input for user to type keystrokes
-key_input = st.text_input("Enter Keystroke:", key="")
-
-# Define voice, speed, and pitch variables (initial values)
-voice = "en"  # English (change for other voices)
-speed = 1.0
-pitch = 1.0
-
-# (Optional) Text pre-processing function (customize for specific needs)
+# Helper function to preprocess text (if needed)
 def preprocess_text(text):
-    # Implement your desired text cleaning or modification logic here
-    return text
-
-# Function to simulate text input to a text-to-speech API
-def process_text(text):
-    global voice, speed, pitch
-
-    # Call pre-processing function (if implemented)
-    preprocessed_text = preprocess_text(text)
+    # Implement any desired text cleaning or formatting here
+    return text.strip()
 
-    # Simulate text input to a text-to-speech API (replace with your chosen API)
-    # This example demonstrates a basic placeholder
-    print(f"Simulating text input to TTS API: {preprocessed_text} (Voice: {voice}, Speed: {speed}, Pitch: {pitch})")
-
-    # You'll need to integrate your chosen text-to-speech API here
+# Helper function to convert text to speech using gTTS
+def text_to_speech(text, language):
+    tts = gTTS(text=text, lang=language, slow=False)
+    file_path = "output.mp3"
+    tts.save(file_path)
+    return file_path
 
 # Streamlit App
-st.title("Text-to-Speech Announcer (Headless)")
+st.title("Text-to-Speech Announcer (Powered by gTTS)")
 
-# User Interface for customization options
-voice_selected = st.selectbox("Voice", ["en", "fr", "es"])  # Add more options
-speed_slider = st.slider("Speaking Speed", min_value=0.5, max_value=2.0, value=1.0)
-pitch_slider = st.slider("Speaking Pitch", min_value=0.5, max_value=2.0, value=1.0)
+# Text input for user to type keystrokes
+key_input = st.text_input("Enter Keystroke:", key="")
 
-# Update variables based on user selections
-voice = voice_selected
-speed = speed_slider
-pitch = pitch_slider
+# User Interface for customization options
+voice_selected = st.selectbox("Language", ["en", "fr", "es"])  # Language codes for gTTS
+speed = st.slider("Speaking Speed (placeholder)", min_value=0.5, max_value=2.0, value=1.0)
+pitch = st.slider("Speaking Pitch (placeholder)", min_value=0.5, max_value=2.0, value=1.0)
 
 # Button to trigger processing of entered text
 if st.button("Announce Keystroke"):
     if key_input:
-        process_text(key_input)
+        processed_text = preprocess_text(key_input)
+        audio_file = text_to_speech(processed_text, voice_selected)
+        
+        # Play the audio file in Streamlit
+        with open(audio_file, "rb") as file:
+            audio_bytes = file.read()
+            b64_audio = base64.b64encode(audio_bytes).decode()
+            audio_player = f"""
+                <audio controls>
+                    <source src="data:audio/mp3;base64,{b64_audio}" type="audio/mp3">
+                    Your browser does not support the audio element.
+                </audio>
+            """
+            st.markdown(audio_player, unsafe_allow_html=True)
+
+        # Clean up the temporary file
+        os.remove(audio_file)
     else:
         st.error("Please enter a keystroke to announce.")
 
-st.write("Enter a keystroke in the text box to synthesize speech.")
+st.write("Enter a keystroke in the text box to synthesize speech.")