File size: 7,641 Bytes
e7eeb32 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 |
import streamlit as st
import base64
import os
from PyPDF2 import PdfReader
import threading
import time
import hashlib
from datetime import datetime
import json
import asyncio
import edge_tts
# Patch asyncio for nested event loops
import nest_asyncio
nest_asyncio.apply()
# Available English voices for Edge TTS
EDGE_TTS_VOICES = [
"en-US-AriaNeural",
"en-US-GuyNeural",
"en-US-JennyNeural",
"en-GB-SoniaNeural",
"en-GB-RyanNeural",
"en-AU-NatashaNeural",
"en-AU-WilliamNeural",
"en-CA-ClaraNeural",
"en-CA-LiamNeural"
]
# Initialize session state for voice selection
if 'tts_voice' not in st.session_state:
st.session_state['tts_voice'] = EDGE_TTS_VOICES[0]
class AudioProcessor:
def __init__(self):
self.cache_dir = "audio_cache"
os.makedirs(self.cache_dir, exist_ok=True)
self.metadata = self._load_metadata()
def _load_metadata(self):
metadata_file = os.path.join(self.cache_dir, "metadata.json")
return json.load(open(metadata_file)) if os.path.exists(metadata_file) else {}
def _save_metadata(self):
metadata_file = os.path.join(self.cache_dir, "metadata.json")
with open(metadata_file, 'w') as f:
json.dump(self.metadata, f)
async def create_audio(self, text, voice='en-US-AriaNeural'):
cache_key = hashlib.md5(f"{text}:{voice}".encode()).hexdigest()
cache_path = os.path.join(self.cache_dir, f"{cache_key}.mp3")
if cache_key in self.metadata and os.path.exists(cache_path):
return open(cache_path, 'rb').read()
# Clean text for speech
text = text.replace("\n", " ").replace("</s>", " ").strip()
if not text:
return None
# Generate audio with edge_tts
communicate = edge_tts.Communicate(text, voice)
await communicate.save(cache_path)
# Update metadata
self.metadata[cache_key] = {
'timestamp': datetime.now().isoformat(),
'text_length': len(text),
'voice': voice
}
self._save_metadata()
return open(cache_path, 'rb').read()
def get_download_link(bin_data, filename, size_mb=None):
b64 = base64.b64encode(bin_data).decode()
size_str = f"({size_mb:.1f} MB)" if size_mb else ""
return f'''
<div class="download-container">
<a href="data:audio/mpeg;base64,{b64}"
download="{filename}" class="download-link">π₯ {filename}</a>
<div class="file-info">{size_str}</div>
</div>
'''
def process_pdf(pdf_file, max_pages, voice, audio_processor):
reader = PdfReader(pdf_file)
total_pages = min(len(reader.pages), max_pages)
texts, audios = [], {}
async def process_page(i, text):
audio_data = await audio_processor.create_audio(text, voice)
audios[i] = audio_data
# Extract text and start audio processing
for i in range(total_pages):
text = reader.pages[i].extract_text()
texts.append(text)
# Process audio in background
threading.Thread(
target=lambda: asyncio.run(process_page(i, text))
).start()
return texts, audios, total_pages
def main():
st.set_page_config(page_title="π PDF to Audio π§", page_icon="π", layout="wide")
# Apply styling
st.markdown("""
<style>
.download-link {
color: #1E90FF;
text-decoration: none;
padding: 8px 12px;
margin: 5px;
border: 1px solid #1E90FF;
border-radius: 5px;
display: inline-block;
transition: all 0.3s ease;
}
.download-link:hover {
background-color: #1E90FF;
color: white;
}
.file-info {
font-size: 0.8em;
color: gray;
margin-top: 4px;
}
</style>
""", unsafe_allow_html=True)
# Initialize processor
audio_processor = AudioProcessor()
# Sidebar settings
st.sidebar.title("π₯ Downloads & Settings")
# Voice selection UI from second app
st.sidebar.markdown("### π€ Voice Settings")
selected_voice = st.sidebar.selectbox(
"π Select TTS Voice:",
options=EDGE_TTS_VOICES,
index=EDGE_TTS_VOICES.index(st.session_state['tts_voice'])
)
st.sidebar.markdown("""
# ποΈ Voice Character Agent Selector π
*Female Voices*:
- πΈ **Aria** β Elegant, creative storytelling
- πΆ **Jenny** β Friendly, conversational
- πΊ **Sonia** β Bold, confident
- π **Natasha** β Sophisticated, mysterious
- π· **Clara** β Cheerful, empathetic
*Male Voices*:
- π **Guy** β Authoritative, versatile
- π οΈ **Ryan** β Approachable, casual
- π» **William** β Classic, scholarly
- π **Liam** β Energetic, engaging
""")
if selected_voice != st.session_state['tts_voice']:
st.session_state['tts_voice'] = selected_voice
st.rerun()
# Main interface
st.markdown("<h1>π PDF to Audio Converter π§</h1>", unsafe_allow_html=True)
col1, col2 = st.columns(2)
with col1:
uploaded_file = st.file_uploader("Choose a PDF file", "pdf")
with col2:
max_pages = st.slider('Select pages to process', min_value=1, max_value=100, value=10)
if uploaded_file:
progress_bar = st.progress(0)
status = st.empty()
with st.spinner('Processing PDF...'):
texts, audios, total_pages = process_pdf(uploaded_file, max_pages, st.session_state['tts_voice'], audio_processor)
for i, text in enumerate(texts):
with st.expander(f"Page {i+1}", expanded=i==0):
st.markdown(text)
# Wait for audio processing
while i not in audios:
time.sleep(0.1)
if audios[i]:
st.audio(audios[i], format='audio/mp3')
# Add download link
if audios[i]:
size_mb = len(audios[i]) / (1024 * 1024)
st.sidebar.markdown(
get_download_link(audios[i], f'page_{i+1}.mp3', size_mb),
unsafe_allow_html=True
)
progress_bar.progress((i + 1) / total_pages)
status.text(f"Processing page {i+1}/{total_pages}")
st.success(f"β
Successfully processed {total_pages} pages!")
# Text to Audio section
st.markdown("### βοΈ Text to Audio")
prompt = st.text_area("Enter text to convert to audio", height=200)
if prompt:
with st.spinner('Converting text to audio...'):
audio_data = asyncio.run(audio_processor.create_audio(prompt, st.session_state['tts_voice']))
if audio_data:
st.audio(audio_data, format='audio/mp3')
size_mb = len(audio_data) / (1024 * 1024)
st.sidebar.markdown("### π΅ Custom Audio")
st.sidebar.markdown(
get_download_link(audio_data, 'custom_text.mp3', size_mb),
unsafe_allow_html=True
)
# Cache management
if st.sidebar.button("Clear Cache"):
for file in os.listdir(audio_processor.cache_dir):
os.remove(os.path.join(audio_processor.cache_dir, file))
audio_processor.metadata = {}
audio_processor._save_metadata()
st.sidebar.success("Cache cleared successfully!")
if __name__ == "__main__":
main() |