Update app.py

app.py

@@ -1,57 +1,45 @@
 import gradio as gr
 from gtts import gTTS
+import numpy as np
+import io
+import base64
 import os
-import tempfile
-from transformers import pipeline
-import torch
 
-# Initialize text-to-audio pipeline for sound generation
-try:
-    device = 0 if torch.cuda.is_available() else -1
-    text_to_audio = pipeline("text-to-audio", model="facebook/musicgen-small", device=device)
-except Exception as e:
-    print(f"Error initializing text-to-audio pipeline: {e}")
-    text_to_audio = None
+def text_to_speech(text, language='en'):
+    tts = gTTS(text=text, lang=language)
+    fp = io.BytesIO()
+    tts.write_to_fp(fp)
+    fp.seek(0)
+    return fp
 
-def generate_speech(text, language):
-    try:
-        tts = gTTS(text=text, lang=language)
-        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as fp:
-            tts.save(fp.name)
-            return gr.Audio(value=fp.name, type="filepath"), "Speech generated successfully"
-    except Exception as e:
-        return None, f"Error in speech generation: {str(e)}"
+def generate_sound(frequency, duration):
+    # Generate a simple sine wave
+    t = np.linspace(0, duration, int(44100 * duration), False)
+    audio = np.sin(2 * np.pi * frequency * t)
+    return (44100, audio.astype(np.float32))
 
-def generate_sound(text):
-    try:
-        if text_to_audio is not None:
-            audio = text_to_audio(text, forward_params={"do_sample": True, "max_new_tokens": 256})
-            with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as fp:
-                audio['audio'].save(fp.name)
-                return gr.Audio(value=fp.name, type="filepath"), "Sound generated successfully"
-        else:
-            return None, "Text-to-audio pipeline not available. Check logs for initialization error."
-    except Exception as e:
-        return None, f"Error in sound generation: {str(e)}"
+def generate_voiceover(text, language):
+    audio_fp = text_to_speech(text, language)
+    return audio_fp
 
-# Gradio interface
-with gr.Blocks() as iface:
-    gr.Markdown("# Lightweight Text-to-Speech and Text-to-Sound Generation Tool")
-    
-    with gr.Tab("Text-to-Speech"):
-        text_input = gr.Textbox(label="Enter text for speech generation")
-        language_input = gr.Dropdown(["en", "es", "fr", "de", "it"], label="Select Language", value="en")
-        speech_button = gr.Button("Generate Speech")
-        speech_output = gr.Audio(label="Generated Speech")
-        speech_message = gr.Textbox(label="Message")
-
-    with gr.Tab("Text-to-Sound"):
-        sound_input = gr.Textbox(label="Enter text description for sound generation")
-        sound_button = gr.Button("Generate Sound")
-        sound_output = gr.Audio(label="Generated Sound")
-        sound_message = gr.Textbox(label="Message")
+def generate_sound_effect(frequency, duration):
+    return generate_sound(frequency, duration)
 
-    speech_button.click(generate_speech, inputs=[text_input, language_input], outputs=[speech_output, speech_message])
-    sound_button.click(generate_sound, inputs=[sound_input], outputs=[sound_output, sound_message])
+# Gradio interface
+iface = gr.Interface(
+    fn=[generate_voiceover, generate_sound_effect],
+    inputs=[
+        gr.Textbox(label="Enter text for speech generation"),
+        gr.Dropdown(choices=["en", "es", "fr", "de", "it"], label="Select Language", value="en"),
+        gr.Number(label="Frequency (Hz)", value=440),
+        gr.Number(label="Duration (seconds)", value=1)
+    ],
+    outputs=[
+        gr.Audio(label="Generated Speech"),
+        gr.Audio(label="Generated Sound Effect")
+    ],
+    title="Simple Text-to-Speech and Sound Generation Tool",
+    description="Generate speech from text and create simple sound effects."
+)
 
 iface.launch()