Edge TTS is a powerful Text-to-Speech (TTS) package for PHP that leverages Microsoft Edge's speech synthesis capabilities. This package allows you to synthesize speech from text, manage voice options, and process audio streams with real-time callbacks through both programmatic and command-line interfaces.
json, curlYou can install Edge TTS via Composer. Run the following command in your terminal:
composer require afaya/edge-tts
This package is a fork of the original afaya/edge-tts package with improvements and fixes. We've downgraded the dependencies to be compatible from PHP 8.1+ to PHP 7.4+ while maintaining the original functionality .
To synthesize speech from text:
php ./vendor/bin/edge-tts edge-tts:synthesize --text "Hello, world!"
To list available voices:
php ./vendor/bin/edge-tts edge-tts:voice-list
<?php
require __DIR__ . '/vendor/autoload.php';
use Afaya\EdgeTTS\Service\EdgeTTS;
// Initialize the EdgeTTS service
$tts = new EdgeTTS();
// Get available voices
$voices = $tts->getVoices();
// var_dump($voices); // array -> use ShortName with the voice name
// Synthesize text with options for voice, rate, volume, and pitch
$tts->synthesize("Hello, world!", 'en-US-AriaNeural', [
'rate' => '0%', // Speech rate (range: -100% to 100%)
'volume' => '0%', // Speech volume (range: -100% to 100%)
'pitch' => '0Hz' // Voice pitch (range: -100Hz to 100Hz)
]);
// Export synthesized audio in different formats
$base64Audio = $tts->toBase64(); // Get audio as base64
$tts->toFile("output"); // Save audio to file
$rawAudio = $tts->toRaw(); // Get raw audio stream
For real-time audio processing:
$tts->synthesizeStream(
"Your text here",
'en-US-AriaNeural',
['rate' => '10%'],
function($chunk) {
// Process each audio chunk in real-time
echo "Received chunk of " . strlen($chunk) . " bytes\n";
// You can stream this directly to output, save incrementally, etc.
}
);
// Get basic audio information
$info = $tts->getAudioInfo();
// Returns: ['size' => bytes, 'format' => 'mp3', 'estimatedDuration' => seconds]
// Get estimated duration
$duration = $tts->getDuration();
// Get size in bytes
$size = $tts->getSizeBytes();
// Get word boundaries with timestamps
$boundaries = $tts->getWordBoundaries();
// Save metadata to file
$tts->saveMetadata('metadata.json');
After synthesizing speech, you can export the audio in various formats:
toBase64(): Returns the audio as a Base64 stringtoFile($path): Saves the audio to a specified filetoRaw(): Returns the raw audio streamtoStream(): Returns a PHP stream resource// Get all available voices
$voices = $tts->getVoices();
// Each voice contains information like:
// - ShortName: The voice identifier to use in synthesis
// - DisplayName: Human-readable voice name
// - LocalName: Localized voice name
// - Gender: Voice gender
// - Locale: Language/region code
The package uses high-quality audio settings by default:
You can customize the audio output format using the outputFormat option. The package supports all formats provided by Microsoft Azure Speech Service:
use Afaya\EdgeTTS\Service\EdgeTTS;
use Afaya\EdgeTTS\Config\Constants;
$tts = new EdgeTTS();
// Use a custom output format
$tts->synthesize("Hello, world!", 'en-US-AriaNeural', [
'outputFormat' => 'audio-48khz-192kbitrate-mono-mp3',
'rate' => '0%',
'pitch' => '0Hz'
]);
$tts->toFile("output"); // Will save as output.mp3
The package supports 36 different audio formats through Constants::OUTPUT_FORMAT:
audio-16khz-32kbitrate-mono-mp3audio-16khz-64kbitrate-mono-mp3audio-16khz-128kbitrate-mono-mp3audio-24khz-48kbitrate-mono-mp3 (default)audio-24khz-96kbitrate-mono-mp3audio-24khz-160kbitrate-mono-mp3audio-48khz-96kbitrate-mono-mp3audio-48khz-192kbitrate-mono-mp3audio-16khz-16bit-32kbps-mono-opusaudio-24khz-16bit-24kbps-mono-opusaudio-24khz-16bit-48kbps-mono-opusogg-16khz-16bit-mono-opusogg-24khz-16bit-mono-opusogg-48khz-16bit-mono-opuswebm-16khz-16bit-mono-opuswebm-24khz-16bit-24kbps-mono-opuswebm-24khz-16bit-mono-opusraw-8khz-8bit-mono-alawraw-8khz-8bit-mono-mulawraw-8khz-16bit-mono-pcmraw-16khz-16bit-mono-pcmraw-16khz-16bit-mono-truesilkraw-22050hz-16bit-mono-pcmraw-24khz-16bit-mono-pcmraw-24khz-16bit-mono-truesilkraw-44100hz-16bit-mono-pcmraw-48khz-16bit-mono-pcmriff-8khz-8bit-mono-alawriff-8khz-8bit-mono-mulawriff-8khz-16bit-mono-pcmriff-22050hz-16bit-mono-pcmriff-24khz-16bit-mono-pcmriff-44100hz-16bit-mono-pcmriff-48khz-16bit-mono-pcmamr-wb-16000hzg722-16khz-64kbpsYou can test all available formats using the included test script:
php test-formats.php
This will test all 36 formats and provide a detailed report showing which formats are compatible with the Edge TTS service.
The package includes comprehensive error handling:
try {
$tts->synthesize("Hello", 'invalid-voice');
} catch (InvalidArgumentException $e) {
echo "Invalid voice: " . $e->getMessage();
} catch (RuntimeException $e) {
echo "Runtime error: " . $e->getMessage();
}
./vendor/bin/phpunit
We welcome contributions! Please read our CONTRIBUTING.md for guidelines on how to contribute to this project.
This project is licensed under the GNU General Public License v3 (GPLv3).
We would like to extend our gratitude to the developers and contributors of the following projects for their inspiration and groundwork:
How can I help you explore Laravel packages today?