Utilizzo - VOICEVOX Core for PHP

<?php

require __DIR__ . '/vendor/autoload.php';

use Revolution\Voicevox\Core\Enums\AccelerationMode;
use Revolution\Voicevox\Core\Onnxruntime;
use Revolution\Voicevox\Core\OpenJtalk;
use Revolution\Voicevox\Core\Synthesizer;
use Revolution\Voicevox\Core\VoiceModelFile;

// Modifica il percorso in base a dove hai installato voicevox_core
$voicevoxCoreDir = getenv('HOME') . '/.local/voicevox_core';
// La versione consigliata di ONNX Runtime è la 1.23.2. Se non corrisponde a quella installata, specifica direttamente il nome del file
$onnxruntimeFilename = $voicevoxCoreDir . '/onnxruntime/lib/' . Onnxruntime::libRecommendedVersionedFilename();
$dictDir = $voicevoxCoreDir . '/dict/open_jtalk_dic_utf_8-1.11';
$vvmPath  = $voicevoxCoreDir . '/models/vvms/0.vvm';

// Testo da sintetizzare e ID di stile
$text    = 'この音声は、ボイスボックスを使用して、出力されています。';
$styleId = 0;
$outPath = './output.wav';

// Inizializzazione
$onnxruntime = Onnxruntime::loadOnce($onnxruntimeFilename);
$openJtalk   = new OpenJtalk($dictDir);
$synthesizer = new Synthesizer($onnxruntime, $openJtalk, AccelerationMode::Auto);

// Carica il modello vocale
$model = VoiceModelFile::open($vvmPath);
$synthesizer->loadVoiceModel($model);

// Sintesi
$audioQuery = $synthesizer->createAudioQuery($text, $styleId);
$wav        = $synthesizer->synthesis($audioQuery, $styleId);

file_put_contents($outPath, $wav);
echo 'Wrote ' . $outPath . PHP_EOL;

php talk.php

$wav = $synthesizer->tts($text, $styleId);
file_put_contents('./output.wav', $wav);

// Usando ttsFromKana
$wav = $synthesizer->ttsFromKana("コノオンセイワ'、ボイスボックスオ'/シヨーシテ'、シュツリョクサレテイマ'ス。", $styleId);
file_put_contents('./output.wav', $wav);

use Revolution\Voicevox\Core\UserDict;
use Revolution\Voicevox\Core\Enums\UserDictWordType;

$userDict = new UserDict();
$uuid = $userDict->addWord(
    surface: 'ボイボ',
    pronunciation: 'ボイボ',
    accentType: 0,
    wordType: UserDictWordType::ProperNoun,
    priority: 10,
);

// Salva il dizionario
$userDict->save('./mydict.json');

// Applica il dizionario a OpenJTalk
$openJtalk->useUserDict($userDict);

$audioQueryJson = $synthesizer->createAudioQuery($text, $styleId);

// Decodifica il JSON e regolalo
$audioQuery = json_decode($audioQueryJson, true);
$audioQuery['speedScale'] = 1.2;   // velocità 1,2x
$audioQuery['pitchScale'] = 0.05;  // pitch leggermente più alto

$wav = $synthesizer->synthesis(json_encode($audioQuery), $styleId);
file_put_contents('./output.wav', $wav);

$accentPhrasesJson = $synthesizer->createAccentPhrases($text, $styleId);
$accentPhrases = json_decode($accentPhrasesJson, true);

// Sovrascrivi pitch delle mora e durata dei fonemi con un altro stile
$updatedJson = $synthesizer->replaceMoraData($accentPhrasesJson, $styleId);

use Revolution\Voicevox\Core\Enums\AccelerationMode;

$synthesizer = new Synthesizer($onnxruntime, $openJtalk, AccelerationMode::Gpu);

// Verifica se la modalità GPU è attiva
if ($synthesizer->isGpuMode()) {
    echo 'In esecuzione in modalità GPU' . PHP_EOL;
}

$model0 = VoiceModelFile::open($voicevoxCoreDir . '/models/vvms/0.vvm');
$model1 = VoiceModelFile::open($voicevoxCoreDir . '/models/vvms/1.vvm');

$synthesizer->loadVoiceModel($model0);
$synthesizer->loadVoiceModel($model1);

// Verifica le informazioni sugli speaker inclusi nel modello
$metas = json_decode($synthesizer->metas(), true);

// Scarica il modello (quando non serve più)
$synthesizer->unloadVoiceModel($model0->id());

use Revolution\Voicevox\Core\Enums\OnExistingVoiceModelId;

$synthesizer->loadVoiceModel($model0, OnExistingVoiceModelId::Skip);

use Revolution\Voicevox\Core\VoicevoxCore;

$audioQuery   = $synthesizer->createAudioQuery($text, $styleId);
$audioFeature = $synthesizer->createAudioFeature($audioQuery, $styleId);

$frameRate        = (new VoicevoxCore)->getFrameRate();
$chunkFrameLength = max(1, (int) round(0.3 * $frameRate));
$totalFrames      = $audioFeature->length();

$stream = fopen('./talk.pcm', 'wb');

for ($start = 0; $start < $totalFrames; $start += $chunkFrameLength) {
    $end = min($start + $chunkFrameLength, $totalFrames);
    fwrite($stream, $audioFeature->slice($start, $end)->render());
}

fclose($stream);