version 5 and version 6 code , browser to mp3
async def generate_chunk_audio(chunks, voice):
audio_segments = []
for i, chunk in enumerate(chunks):
print(f"🎤 Processing chunk {i+1}/{len(chunks)} ({len(chunk)} chars)")
chunk_audio = AudioSegment.empty()
# 🔸 Split the chunk into sentences by full stop
sentences = [s.strip() for s in chunk.split('.') if s.strip()]
for j, sentence in enumerate(sentences):
sentence += '.' # add the period back
sentence_path = os.path.join(temp_dir, f"chunk_{i}_{j}.mp3")
try:
communicate = edge_tts.Communicate(sentence, voice)
await communicate.save(sentence_path)
audio = AudioSegment.from_file(sentence_path)
chunk_audio += audio
# ⏸ Add 1 second pause after each sentence
chunk_audio += AudioSegment.silent(duration=1000)
except Exception as e:
print(f"⚠️ Failed sentence {j} in chunk {i}: {e}")
audio_segments.append(chunk_audio)
return audio_segments
----------
# ✅ Install required packages
!pip install edge-tts pydub --quiet
!apt install ffmpeg --quiet
import requests
from bs4 import BeautifulSoup
import re
import asyncio
import edge_tts
from pydub import AudioSegment
import os
import shutil
# 📌 Config
url = "https://prabodhadevotee.blogspot.com/2025/04/pss-visitor-recording.html"
start_phrase = "శ్లోకం 11"
end_phrase = "గాలికి త్రోయబడదు"
voice = "te-IN-ShrutiNeural" # Telugu Female
chunk_size = 3000
output_path = "/content/smallBook_output_female_final.mp3"
temp_dir = "/content/chunks"
os.makedirs(temp_dir, exist_ok=True)
# 🔍 Step 1: Extract blog text between start and end phrases
response = requests.get(url)
soup = BeautifulSoup(response.content, "html.parser")
full_text = soup.get_text()
pattern = re.compile(f"{re.escape(start_phrase)}(.*?){re.escape(end_phrase)}", re.DOTALL)
match = pattern.search(full_text)
if match:
selected_text = start_phrase + match.group(1) + end_phrase
else:
raise Exception("Text not found between start and end phrases.")
selected_text = re.sub(r'\s+', ' ', selected_text).strip()
# 🔠 Step 2: Split text into manageable chunks
def split_text(text, max_len):
words = text.split()
chunks, current = [], ""
for word in words:
if len(current) + len(word) + 1 <= max_len:
current += (" " + word if current else word)
else:
chunks.append(current)
current = word
if current:
chunks.append(current)
return chunks
chunks = split_text(selected_text, chunk_size)
print(f"🔍 Total Chunks: {len(chunks)}")
# 🎤 Step 3: Generate MP3 from each chunk and insert 1s silence after each sentence (excluding last)
async def generate_chunk_audio(chunks, voice):
audio_segments = []
for i, chunk in enumerate(chunks):
print(f"🎤 Generating chunk {i+1}/{len(chunks)} ({len(chunk)} chars)")
sentences = re.split(r'(?<=\.)\s+', chunk) # split on ". "
chunk_audio = AudioSegment.empty()
for j, sentence in enumerate(sentences):
if sentence.strip():
chunk_path = os.path.join(temp_dir, f"chunk_{i}_{j}.mp3")
try:
communicate = edge_tts.Communicate(sentence, voice)
await communicate.save(chunk_path)
audio = AudioSegment.from_file(chunk_path)
chunk_audio += audio
if j < len(sentences) - 1: # 👈 Avoid extra silence at end
chunk_audio += AudioSegment.silent(duration=1000)
except Exception as e:
print(f"⚠️ Chunk {i}, Sentence {j} failed: {e}")
audio_segments.append(chunk_audio)
return audio_segments
# 🎧 Step 4: Combine all chunks and clean up
async def process():
audio_segments = await generate_chunk_audio(chunks, voice)
if not audio_segments:
print("❌ No audio segments generated.")
return
final_audio = sum(audio_segments[1:], audio_segments[0])
final_audio.export(output_path, format="mp3")
print(f"✅ Final MP3 saved at: {output_path}")
print(f"📏 Final Duration: {round(len(final_audio) / 60000, 2)} minutes")
# 🧹 Clean up
shutil.rmtree(temp_dir)
# 🚀 Run
await process()
