Merge pull request #52051 from Lukasdotcom/feat/text-to-speech

feat(TaskProcessing): Add TextToSpeech provider

lib/composer/composer/autoload_classmap.php

@@ -837,6 +837,7 @@ return array(
     'OCP\\TaskProcessing\\TaskTypes\\ContextWrite' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/ContextWrite.php',
     'OCP\\TaskProcessing\\TaskTypes\\GenerateEmoji' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/GenerateEmoji.php',
     'OCP\\TaskProcessing\\TaskTypes\\TextToImage' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/TextToImage.php',
+    'OCP\\TaskProcessing\\TaskTypes\\TextToSpeech' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/TextToSpeech.php',
     'OCP\\TaskProcessing\\TaskTypes\\TextToText' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/TextToText.php',
     'OCP\\TaskProcessing\\TaskTypes\\TextToTextChangeTone' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/TextToTextChangeTone.php',
     'OCP\\TaskProcessing\\TaskTypes\\TextToTextChat' => $baseDir . '/lib/public/TaskProcessing/TaskTypes/TextToTextChat.php',

lib/composer/composer/autoload_static.php

@@ -878,6 +878,7 @@ class ComposerStaticInit749170dad3f5e7f9ca158f5a9f04f6a2
         'OCP\\TaskProcessing\\TaskTypes\\ContextWrite' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/ContextWrite.php',
         'OCP\\TaskProcessing\\TaskTypes\\GenerateEmoji' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/GenerateEmoji.php',
         'OCP\\TaskProcessing\\TaskTypes\\TextToImage' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/TextToImage.php',
+        'OCP\\TaskProcessing\\TaskTypes\\TextToSpeech' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/TextToSpeech.php',
         'OCP\\TaskProcessing\\TaskTypes\\TextToText' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/TextToText.php',
         'OCP\\TaskProcessing\\TaskTypes\\TextToTextChangeTone' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/TextToTextChangeTone.php',
         'OCP\\TaskProcessing\\TaskTypes\\TextToTextChat' => __DIR__ . '/../../..' . '/lib/public/TaskProcessing/TaskTypes/TextToTextChat.php',

lib/private/TaskProcessing/Manager.php

@@ -584,6 +584,7 @@ class Manager implements IManager {
 			\OCP\TaskProcessing\TaskTypes\TextToTextChatWithTools::ID => \OCP\Server::get(\OCP\TaskProcessing\TaskTypes\TextToTextChatWithTools::class),
 			\OCP\TaskProcessing\TaskTypes\ContextAgentInteraction::ID => \OCP\Server::get(\OCP\TaskProcessing\TaskTypes\ContextAgentInteraction::class),
 			\OCP\TaskProcessing\TaskTypes\TextToTextProofread::ID => \OCP\Server::get(\OCP\TaskProcessing\TaskTypes\TextToTextProofread::class),
+			\OCP\TaskProcessing\TaskTypes\TextToSpeech::ID => \OCP\Server::get(\OCP\TaskProcessing\TaskTypes\TextToSpeech::class),
 		];
 
 		foreach ($context->getTaskProcessingTaskTypes() as $providerServiceRegistration) {

lib/public/TaskProcessing/TaskTypes/TextToSpeech.php

@@ -0,0 +1,92 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * SPDX-FileCopyrightText: 2024 Nextcloud GmbH and Nextcloud contributors
+ * SPDX-License-Identifier: AGPL-3.0-or-later
+ */
+
+namespace OCP\TaskProcessing\TaskTypes;
+
+use OCP\IL10N;
+use OCP\L10N\IFactory;
+use OCP\TaskProcessing\EShapeType;
+use OCP\TaskProcessing\ITaskType;
+use OCP\TaskProcessing\ShapeDescriptor;
+
+/**
+ * This is the task processing task type for speech generation
+ * @since 32.0.0
+ */
+class TextToSpeech implements ITaskType {
+	/**
+	 * @since 32.0.0
+	 */
+	public const ID = 'core:text2speech';
+
+	private IL10N $l;
+
+	/**
+	 * @param IFactory $l10nFactory
+	 * @since 32.0.0
+	 */
+	public function __construct(
+		IFactory $l10nFactory,
+	) {
+		$this->l = $l10nFactory->get('lib');
+	}
+
+
+	/**
+	 * @inheritDoc
+	 * @since 32.0.0
+	 */
+	public function getName(): string {
+		return $this->l->t('Generate speech');
+	}
+
+	/**
+	 * @inheritDoc
+	 * @since 32.0.0
+	 */
+	public function getDescription(): string {
+		return $this->l->t('Generate speech from a transcript');
+	}
+
+	/**
+	 * @return string
+	 * @since 32.0.0
+	 */
+	public function getId(): string {
+		return self::ID;
+	}
+
+	/**
+	 * @return ShapeDescriptor[]
+	 * @since 32.0.0
+	 */
+	public function getInputShape(): array {
+		return [
+			'input' => new ShapeDescriptor(
+				$this->l->t('Prompt'),
+				$this->l->t('Write transcript that you want the assistant to generate speech from'),
+				EShapeType::Text
+			),
+		];
+	}
+
+	/**
+	 * @return ShapeDescriptor[]
+	 * @since 32.0.0
+	 */
+	public function getOutputShape(): array {
+		return [
+			'speech' => new ShapeDescriptor(
+				$this->l->t('Output speech'),
+				$this->l->t('The generated speech'),
+				EShapeType::Audio
+			),
+		];
+	}
+}