grok, txt to mp3 worked without google with some limitations

 

code:

!pip install gTTS pydub

from gtts import gTTS

from pydub import AudioSegment

import os

# Paths to the files

text_file_path = '/content/pssvisitor.txt'

output_file_path = '/content/pssvisitor_.mp3'

# Verify the file existence

if not os.path.exists(text_file_path):

    raise FileNotFoundError(f"Text file {text_file_path} not found")

# Load text from the file

with open(text_file_path, 'r', encoding='utf-8') as file:

    text = file.read().strip()

# Function to split text into chunks based on byte size

def split_text(text, max_bytes=5000):

    chunks = []

    current_chunk = ""

    current_chunk_bytes = 0

    for char in text:

        char_bytes = len(char.encode('utf-8'))

        if current_chunk_bytes + char_bytes > max_bytes:

            chunks.append(current_chunk)

            current_chunk = char

            current_chunk_bytes = char_bytes

        else:

            current_chunk += char

            current_chunk_bytes += char_bytes

    if current_chunk:

        chunks.append(current_chunk)

    return chunks

# Configure voice and speed

language = 'te'  # Telugu ('te'), can switch to 'en' for English

voice_gender = 'male'  # gTTS does not support gender selection; defaults to neutral/male-like

speaking_rate = 1.0  # Normal speed: 0.5 (slow), 1.0 (normal), 1.5 (fast)

# Function to synthesize speech for each chunk

def synthesize_speech(text_chunk, temp_file_path):

    tts = gTTS(text=text_chunk, lang=language, slow=(speaking_rate < 1.0))

    tts.save(temp_file_path)

# Process text and combine audio files

text_chunks = split_text(text)

temp_files = []

for i, chunk in enumerate(text_chunks):

    temp_file_path = f'/content/temp_chunk_{i}.mp3'

    temp_files.append(temp_file_path)

    synthesize_speech(chunk, temp_file_path)

# Combine audio files

combined = AudioSegment.empty()

for temp_file in temp_files:

    audio_segment = AudioSegment.from_mp3(temp_file)

    combined += audio_segment

# Export the combined audio to the final output file

combined.export(output_file_path, format='mp3')

print(f'Audio content written to file "{output_file_path}"')

# Clean up temporary files

for temp_file in temp_files:

    os.remove(temp_file)

only list of lanugaes and support speed and mode(male/female) information fro above?

ChatGPT said:

The gTTS (Google Text-to-Speech) library is a wrapper over the Google Translate TTS API, and has the following characteristics in terms of languages, speed, and voice types:

---

✅ Supported Languages (with codes)

Here are some commonly supported languages and their codes used with gTTS(lang=...):

Language	Code	TLD Option (Accent)
English	en	com, co.uk, co.in
Hindi	hi	-
Telugu	te	-
Tamil	ta	-
Kannada	kn	-
Malayalam	ml	-
Gujarati	gu	-
Bengali	bn	-
Marathi	mr	-
Urdu	ur	-
Spanish	es	com.mx, es
French	fr	fr, ca
German	de	-
Japanese	ja	-
Chinese (Mandarin)	zh-CN	-
Russian	ru	-

To get the complete list via code:

python
from gtts.lang import tts_langs
print(tts_langs())

---

🐢 Speed Support

gTTS supports two speed modes only:

• slow=False → Normal speed (default)

• slow=True → Slower voice output (about 60% of original)

No support for precise or dynamic speed control.

---

👤 Voice Mode (Male/Female)

• Only one voice (Female) is supported per language.

• You cannot switch between male/female voices using gTTS.

• Voice tone and accent might vary slightly with tld (region) but still female.

---

🔁 Summary Table

Feature	Options / Notes
Languages	~100+ (English, Telugu, Hindi, etc.)
Speed	slow=True (slow), slow=False (normal)
Gender	Female only
Accent control	Use tld='com', 'co.uk', 'co.in', etc.