Update app.py

app.py

@@ -1,8 +1,8 @@
 import gradio as gr
 import requests
 import tempfile
-import pyttsx3
 from faster_whisper import WhisperModel
+from gtts import gTTS
 
 # === Config ===
 GROQ_API_KEY = "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"
@@ -12,15 +12,6 @@ GROQ_API_URL = "https://api.groq.com/openai/v1/chat/completions"
 # === Init Whisper ===
 whisper = WhisperModel("base", device="cpu", compute_type="int8")
 
-# === Init TTS (offline) ===
-tts_engine = pyttsx3.init()
-
-def speak_to_file(text, output_file="output.wav"):
-    """Save TTS to a file using pyttsx3."""
-    tts_engine.save_to_file(text, output_file)
-    tts_engine.runAndWait()
-    return output_file
-
 def process_audio(audio_file):
     # 1. Speech to Text
     segments, _ = whisper.transcribe(audio_file)
@@ -43,9 +34,10 @@ def process_audio(audio_file):
 
     reply = response.json()["choices"][0]["message"]["content"]
 
-    # 3. TTS to file
-    audio_output = tempfile.NamedTemporaryFile(suffix=".wav", delete=False)
-    speak_to_file(reply, audio_output.name)
+    # 3. TTS using gTTS (generates .mp3)
+    tts = gTTS(reply)
+    audio_output = tempfile.NamedTemporaryFile(suffix=".mp3", delete=False)
+    tts.save(audio_output.name)
 
     return reply, audio_output.name
 
@@ -56,8 +48,8 @@ iface = gr.Interface(
         gr.Textbox(label="🧠 Groq Response"),
         gr.Audio(label="🔊 AI Voice Reply")
     ],
-    title="🗣️ Voice AI Assistant (Groq + Whisper + Offline TTS)",
-    description="Use your mic. Whisper transcribes, Groq replies, voice response from pyttsx3 (offline).",
+    title="🗣️ Voice AI Assistant (Groq + Whisper + gTTS)",
+    description="🎙️ Whisper for STT, Groq for response, gTTS for voice output. Fully Hugging Face compatible.",
     live=True
 )