VFN-TTS 1.0.0

Pure PHP Multilingual Natural Text-to-Speech

Developer: OmarRizk
An original work by OmarRizk, developed since 2019.
Copyright (c) 2019-2026 OmarRizk.

VFN-TTS is a pure PHP multilingual Text-to-Speech engine with Natural
speech generation, model can naturally
understand emojis and reactions.

No external API or Python runtime is required.

FEATURES

• Pure PHP runtime.
• No external API required.
• No Python required.
• Multilingual frontend support:
  • Arabic
  • English
  • French
• Automatic language detection.
• Natural emoji/reaction support.
• Automatic voice model discovery.

SUPPORTED LANGUAGES

Arabic:

array('language' => 'ar')

English:

array('language' => 'en')

French:

array('language' => 'fr')

Automatic detection:

array('language' => 'auto')

Recommended mode:

language=auto

SUPPORTED REACTION EMOJI

😂 🤣 😆 😄 😁 😅 🤭 🤔 🙄 😮 😲 😱 😢 😭 😔 😮‍💨 😌 😴 🤧 😷

The model understands emojis and user interactions.

FILES

VFN_TTS/VFN_TTS.php
    Official VFN-TTS.

VFN_TTS/voices/orizk.orizk
    Packed ORIZK voice model.
    Downloaded separately from the official model link below.

example.php
    Safe CLI/web usage example.

web.php
    Demo browser interface.

composer.json
    Composer package metadata.

MODEL DOWNLOAD

The official ORIZK voice model is distributed separately.

Model file:

orizk.orizk

Download:

https://drive.google.com/file/d/1wLDyO5lhEE5_j-xQmd8cxqdpMP1O3Ct_/view?usp=sharing

After downloading, place the model file in:

VFN_TTS/voices/orizk.orizk

The expected structure is:

VFN_TTS/
├── VFN_TTS.php
└── voices/
    └── orizk.orizk

VFN-TTS automatically discovers compatible .orizk voice models stored
inside the VFN_TTS/voices/ directory.

No model conversion, external API, or Python is required.

BASIC USAGE

require __DIR__ . '/VFN_TTS/VFN_TTS.php';

$tts = new FvnTTS(__DIR__ . '/VFN_TTS/voices');

$tts->voice('orizk');

$tts->save(
    'Hello world 😂',
    __DIR__ . '/out.wav',
    array(
        'language' => 'auto'
    )
);

WEB INTERFACE

Open:

web.php

in a browser.

Enter the text to synthesize, choose one of:

• Auto
• Arabic
• English
• French

and press the Generate button.

Generated WAV files are stored in:

generated/

beside web.php.

VOICE DISCOVERY

Additional VFN-TTS voice models can be installed by copying another:

.orizk

file into:

VFN_TTS/voices/

Compatible voice models are discovered automatically.

LICENSE / COPYRIGHT

VFN-TTS is an original work by OmarRizk and VFN Media Lab.

Developed since 2019.

Copyright (c) 2019-2026 OmarRizk.
All rights reserved.