bestmomo/laravel-edge-tts

Edge-tts-php integration for Laravel

0.1.3 2025-10-05 11:47 UTC

Requires

• php: ^8.2
• afaya/edge-tts: ^1.0
• illuminate/support: ^10.0|^11.0|^12.0

Requires (Dev)

• orchestra/testbench: ^8.0|^9.0|^10.0
• phpunit/phpunit: ^10.0|^11.0

MIT

• Bestmomo

README

A simple yet powerful Laravel package for integrating Microsoft Edge Text-to-Speech (TTS) into your applications. It features audio streaming, caching, abstraction, and security controls.

This package relies on the excellent work of andresayac/edge-tts-php to interface with the Microsoft Edge TTS API. All voice support and core synthesis features are inherited from this library.

Installation

1. Via Composer

Add the package to your Laravel project:

composer require bestmomo/laravel-edge-tts

2. File Publication

After installation, publish the configuration file:

php artisan vendor:publish --tag=edge-tts-config

This copies the config/edge-tts.php file, where you can define the default voice and security settings (middlewares).

Configuration

The config/edge-tts.php File

The configuration file allows you to customize key aspects of the package:

return [
    // Default voice used if none is specified in the request
    'default_voice' => env('EDGE_TTS_DEFAULT_VOICE', 'fr-FR-DeniseNeural'),

    // Middlewares applied to the streaming route
    'middleware' => [
        'web',          // Required for session (and therefore 'auth') to work
        'auth',         // Require the user to be logged in
        'throttle:60,1',// Limit to 60 requests per minute
    ],
    
    // Audio caching parameters
    'cache' => [
        'enabled' => env('EDGE_TTS_CACHE_ENABLED', true),
        // Laravel storage disk to use for MP3 files (defaults to 'local')
        'disk' => env('EDGE_TTS_CACHE_DISK', 'local'),
        'lifetime' => env('EDGE_TTS_CACHE_LIFETIME', null), // Cache lifetime in minutes (null = indefinite)
    ],

    // Enable Call Logging
    'enable_call_logging' => env('EDGE_TTS_LOG_CALLS', false),
];

Usage

The package registers an abstraction contract (Interface) and a Facade, allowing you to use it easily anywhere in your application.

1. Using the Contract (Dependency Injection)

The recommended approach is to inject the TtsSynthesizer contract:

use Bestmomo\LaravelEdgeTts\Contracts\TtsSynthesizer;

class MyService
{
    protected $tts;

    public function __construct(TtsSynthesizer $tts)
    {
        $this->tts = $tts;
    }

    public function generateAudio()
    {
        // 1. Save the audio to a file:
        $filePath = $this->tts->toFile(
            'This is a file save test.', 
            'en-US-JennyNeural', 
            storage_path('app/audio/test_jenny')
        );

        // 2. Get the audio as a Base64 string:
        $base64 = $this->tts->toBase64('This is Base64 encoded audio.', 'en-US-JennyNeural');
    }
}

1.1 Using the Facade

You can also use the EdgeTts facade to access the TTS synthesizer:

use Bestmomo\LaravelEdgeTts\Facades\EdgeTts;

class MyService
{
    public function generateAudio()
    {
        // 1. Save the audio to a file:
        $filePath = EdgeTts::toFile(
            'This is a file save test.', 
            'en-US-JennyNeural', 
            storage_path('app/audio/test_jenny')
        );

        // 2. Get the audio as a Base64 string:
        $base64 = EdgeTts::toBase64('This is Base64 encoded audio.', 'en-US-JennyNeural');
    }
}

2. Using the Streaming Route

The core of the package is the audio streaming route, which handles caching and security:

Route URL:

/edge-tts/stream

Route Name:

edge-tts.stream

You can use it directly in your JavaScript code for an <audio> element:

const text = "Hello world!";
const streamUrl = `/edge-tts/stream?text=${encodeURIComponent(text)}&voice=en-US-JennyNeural`;

document.getElementById('audioPlayer').src = streamUrl;
document.getElementById('audioPlayer').play();

Query Parameters:

• text the text to synthesize (can be SSML). (Required)
• voice the voice to use. (default defined in config)
• rate speech rate (e.g., +10%, -5%). Default 0%
• volume voice volume (e.g., +50%, -10%). Default 0%
• pitch voice pitch (e.g., +5Hz, -2Hz). Default 0Hz

SSML Note: If the text parameter starts with <speak, it will be interpreted as SSML (Speech Synthesis Markup Language), allowing for advanced control over pauses, emphasis, and emotion. The simple modulation parameters (rate, volume, pitch) will be ignored in favor of the SSML tags. For example:

<speak version="1.0" xml:lang="en-US">
    <voice xml:lang="en-US" xml:gender="Female" name="en-US-JennyNeural">
        <prosody rate="+10%" volume="+50%" pitch="+5Hz">
            Hello world!
        </prosody>
    </voice>
</speak>

3. Using the Blade Directive

For even simpler integration into your Blade templates, the package provides an @edge_tts directive. It generates an <audio> element configured for streaming.

Basic Usage:

{{-- Synthesize text with the default voice --}}
@edge_tts('text' => 'Hello world!')

Usage with Specific Voice and Parameters:

The directive accepts the same parameters as the streaming route, passed as arguments: text, voice, rate, volume, and pitch.

{{-- Synthesize in French with specific options --}}
@edge_tts([
    'text' => 'Bonjour le monde !', 
    'voice' => 'fr-FR-DeniseNeural',
    'rate' => '+10%'
])

This makes it incredibly easy to embed synthesized audio directly into your views.

Key Features

Caching

The package uses the edge-tts.cache configuration to cache generated audio files. If the exact same combination of (text, voice, options) is requested multiple times, the MP3 file is served instantly from local storage, reducing latency and external API requests.

Enable Call Logging

The enable_call_logging option in the config/edge-tts.php file allows you to log every TTS API call to the Laravel log. This can be useful for debugging purposes or to monitor the usage of the TTS service.

To enable it, set the EDGE_TTS_LOG_CALLS environment variable to true in your .env file:

EDGE_TTS_LOG_CALLS=true

Cache Pruning

The package includes an Artisan command to automatically delete old cache files:

php artisan edge-tts:cache-prune

By default, the command deletes files older than 90 days. You can customize this duration using the --days option:

// Delete files older than 30 days
php artisan edge-tts:cache-prune --days=30

To schedule it, add the following to your routes/console.php file:

use Illuminate\Support\Facades\Schedule;

Schedule::command('edge-tts:cache-prune --days=60')->daily();

Abstraction

The package utilizes a TtsSynthesizer contract. If you ever need to change the TTS provider (e.g., to Google Cloud TTS or Amazon Polly), you only need to create a new adapter implementing this contract and update the Service Provider binding, without changing your application's business logic.

Demo (Development Environment)

You can access the demonstration interface (available only in local, staging, or testing environments):

/edge-tts/demo

Contributing

Contributions are welcome! Please open an issue or submit a pull request with your improvements.

License

This package is open-source software licensed under the MIT license.