Эх сурвалжийг харах

Initial commit for Speech

FusionPBX 1 жил өмнө
parent
commit
bde1ac9632

+ 40 - 0
app_config.php

@@ -0,0 +1,40 @@
+<?php
+
+	//application details
+		$apps[$x]['name'] = 'Speech';
+		$apps[$x]['uuid'] = 'dff26a0d-9439-4743-b787-765f1dd5ebfb';
+		$apps[$x]['category'] = 'API';
+		$apps[$x]['subcategory'] = '';
+		$apps[$x]['version'] = '1.0';
+		$apps[$x]['license'] = 'Mozilla Public License 1.1';
+		$apps[$x]['url'] = 'http://www.fusionpbx.com';
+		$apps[$x]['description']['en-us'] = 'Artificial Intelligence';
+
+	//default settings
+		$y=0;
+		$apps[$x]['default_settings'][$y]['default_setting_uuid'] = "c4751875-011d-4181-ba9a-8978ffd7497d";
+		$apps[$x]['default_settings'][$y]['default_setting_category'] = "speech";
+		$apps[$x]['default_settings'][$y]['default_setting_subcategory'] = "enabled";
+		$apps[$x]['default_settings'][$y]['default_setting_name'] = "boolean";
+		$apps[$x]['default_settings'][$y]['default_setting_value'] = "true";
+		$apps[$x]['default_settings'][$y]['default_setting_enabled'] = "false";
+		$apps[$x]['default_settings'][$y]['default_setting_description'] = "Text to Speech API enabled.";
+		$y++;
+		$apps[$x]['default_settings'][$y]['default_setting_uuid'] = "e7a77c36-92d1-4fb6-9db3-4f62866dbaf2";
+		$apps[$x]['default_settings'][$y]['default_setting_category'] = "speech";
+		$apps[$x]['default_settings'][$y]['default_setting_subcategory'] = "engine";
+		$apps[$x]['default_settings'][$y]['default_setting_name'] = "text";
+		$apps[$x]['default_settings'][$y]['default_setting_value'] = "openai";
+		$apps[$x]['default_settings'][$y]['default_setting_enabled'] = "false";
+		$apps[$x]['default_settings'][$y]['default_setting_description'] = "Text to Speech API engine.";
+		$y++;
+		$apps[$x]['default_settings'][$y]['default_setting_uuid'] = "eced068b-db30-4257-aa7c-6e2659271e4b";
+		$apps[$x]['default_settings'][$y]['default_setting_category'] = "speech";
+		$apps[$x]['default_settings'][$y]['default_setting_subcategory'] = "api_key";
+		$apps[$x]['default_settings'][$y]['default_setting_name'] = "text";
+		$apps[$x]['default_settings'][$y]['default_setting_value'] = "";
+		$apps[$x]['default_settings'][$y]['default_setting_enabled'] = "false";
+		$apps[$x]['default_settings'][$y]['default_setting_description'] = "Text to Speech API Key";
+		$y++;
+
+?>

+ 156 - 0
resources/classes/speech.php

@@ -0,0 +1,156 @@
+<?php
+
+/**
+ * audio class
+ *
+ * @method null download
+ */
+if (!class_exists('speech')) {
+	class speech {
+
+		/**
+		 * declare private variables
+		 */
+		private $api_key;
+
+		/** @var string $engine */
+		private $engine;
+
+		/** @var template_engine $object */
+		private $speech_object;
+
+		private $settings;
+
+		public $audio_path;
+		public $audio_filename;
+		public $audio_format;
+		public $audio_model;
+		public $audio_voice;
+		public $audio_language;
+		public $audio_message;
+
+		/**
+		 * called when the object is created
+		 */
+		public function __construct(settings $settings = null) {
+			//make the setting object
+			if ($settings === null) {
+				$settings = new settings();
+			}
+
+			//add the settings object to the class
+			$this->settings = $settings;
+
+			//build the setting object and get the recording path
+			$this->api_key = $settings->get('speech', 'api_key');
+			$this->engine = $settings->get('speech', 'engine');
+		}
+
+		/**
+		 * get_voices - get the list voices
+		 */
+		public function get_voices() : array {
+
+			//set the class interface to use the _template suffix
+			$classname = 'speech_'.$this->engine;
+
+			//create the object
+			$object = new $classname($this->settings);
+
+			//return the voices array
+			return $object->get_voices();
+		}
+
+		/**
+		 * get_voices - get the list voices
+		 */
+		public function get_models() : array {
+
+			//set the class interface to use the _template suffix
+			$classname = 'speech_'.$this->engine;
+
+			//create the object
+			$object = new $classname($this->settings);
+
+			//return the voices array
+			return $object->get_models();
+		}
+
+		/**
+		 * is_translate_enabled - get whether the engine can do translations
+		 */
+		public function is_translate_enabled() : bool {
+
+			//set the class interface to use the _template suffix
+			$classname = 'speech_'.$this->engine;
+
+			//create the object
+			$object = new $classname($this->settings);
+
+			//return the translate_enabled
+			return $object->is_translate_enabled();
+		}
+
+		/**
+		 * is_language_enabled - get whether the engine allows to set the language
+		 */
+		public function is_language_enabled() : bool {
+
+			//set the class interface to use the _template suffix
+			$classname = 'speech_'.$this->engine;
+
+			//create the object
+			$object = new $classname($this->settings);
+
+			//return the language_enabled
+			return $object->is_language_enabled();
+		}
+
+		/**
+		 * get_languages - get the list languages
+		 */
+		public function get_languages() : array {
+
+			//set the class interface to use the _template suffix
+			$classname = 'speech_'.$this->engine;
+
+			//create the object
+			$object = new $classname($this->settings);
+
+			//return the languages array
+			return $object->get_languages();
+		}
+
+		/**
+		 * speech - text to speech
+		 */
+		public function speech() {
+			if (!empty($this->engine)) {
+				//set the class interface to use the _template suffix
+				$class_name = 'speech_'.$this->engine;
+
+				//create the object
+				$object = new $class_name($this->settings);
+
+				//ensure the class has implemented the speech_interface interface
+				if ($object instanceof speech_interface) {
+					$object->set_path($this->audio_path);
+					$object->set_filename($this->audio_filename);
+					$object->set_format($this->audio_format);
+					$object->set_voice($this->audio_voice);
+					//$object->set_model($this->audio_model);
+					//$object->set_language($this->audio_language);
+					//$object->set_translate($this->audio_translate);
+					$object->set_message($this->audio_message);
+					$object->speech();
+				}
+				else {
+					return false;
+				}
+			}
+		}
+
+	}
+}
+
+?>

+ 236 - 0
resources/classes/speech_elevenlabs.php

@@ -0,0 +1,236 @@
+<?php
+
+
+/**
+ * ai_elevenlabs class
+ *
+ */
+class speech_elevenlabs implements speech_interface {
+
+	private $voice;
+	private $path;
+	private $message;
+	private $format;
+	private $filename;
+	private $languages;
+	private $api_key;
+	private $model;
+
+	public function __construct($settings) {
+		$this->voice = "";
+		$this->path = "";
+		$this->message = "";
+		$this->format = "";
+		$this->filename = "";
+		//build the setting object and get the recording path
+		$this->api_key = $settings->get('speech', 'api_key');
+	}
+
+	public function set_filename(string $audio_filename) {
+		$this->filename = $audio_filename;
+	}
+
+	public function set_format(string $audio_format) {
+		$this->format = $audio_format;
+	}
+
+	public function set_message(string $audio_message) {
+		$this->message = $audio_message;
+	}
+
+	public function set_path(string $audio_path) {
+		$this->path = $audio_path;
+	}
+
+	public function set_voice(string $audio_voice) {
+		$this->voice = $audio_voice;
+	}
+
+	public function speech(): bool {
+
+		//get the model automatically
+		$model_id = $this->get_model();
+
+		// set the request URL
+		$url = 'https://api.elevenlabs.io/v1/text-to-speech/' . $this->voice;
+
+		// set the request headers
+		$headers[] = 'Content-Type: application/json';
+		$headers[] = 'xi-api-key: '.$this->api_key;
+
+		// set the http data
+		$data['model_id'] = $model_id;
+		$data['text'] = $this->message;
+		//$data['pronunciation_dictionary_locators'][0]['pronunciation_dictionary_id'];
+		//$data['pronunciation_dictionary_locators'][0]['version_id'];
+		$data['voice_settings']['similarity_boost'] = 1;
+		$data['voice_settings']['stability'] = 1;
+		$data['voice_settings']['style'] = 0;
+		$data['voice_settings']['use_speaker_boost'] = 'true';
+
+		// initialize curl handle
+		$ch = curl_init($url);
+
+		// set the curl options
+		curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+		curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
+		curl_setopt($ch, CURLOPT_POST, true);
+		curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($data));
+
+		// run the curl request and get the response
+		$response = curl_exec($ch);
+
+		// get the errors
+		$error = curl_error($ch);
+
+		// get the http code
+		$http_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
+
+		// close the handle
+		curl_close($ch);
+
+		// show the result when there is an error
+		if ($http_code != 200) {
+			echo "error ".$error."\n";
+			echo "http_code ".$http_code."\n";
+			if (strlen($response) < 500) {
+				view_array(json_decode($response, true));
+			}
+			exit;
+		}
+
+		// save the audio file
+		if ($http_code == 200) {
+			file_put_contents($this->path.'/'.$this->filename, $response);
+			return true;
+		}
+		return false;
+
+		//$curl = new curl('https://api.elevenlabs.io/v1/text-to-speech/' . $this->voice);
+		//$response = $curl->set_headers($headers)->post(json_encode($data));
+		//$error = $curl->get_error();
+		//$http_code = $curl->get_http_code();
+		//if ($curl->get_http_code() == 200) {
+		//save the audio
+		//if ($http_code == 200) {
+		//	file_put_contents($this->path . '/' . $this->filename, $response);
+		//	return true;
+		//}
+		//return false;
+	}
+
+	public function is_language_enabled(): bool {
+		return false;
+	}
+
+	public function is_model_enabled(): bool {
+		return false;
+	}
+
+	public function get_languages(): array {
+		return ['en' => 'English'];
+	}
+
+	public function get_voices(): array {
+		$return_value = [];
+		$url = 'https://api.elevenlabs.io/v1/voices';
+		$headers = [
+			'Content-Type: application/json',
+			"xi-api-key: $this->api_key",
+		];
+		$curl = curl_init();
+		curl_setopt_array($curl, [
+			CURLOPT_URL => $url,
+			CURLOPT_RETURNTRANSFER => true,
+			CURLOPT_ENCODING => "",
+			CURLOPT_MAXREDIRS => 10,
+			CURLOPT_TIMEOUT => 30,
+			CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
+			CURLOPT_CUSTOMREQUEST => "GET",
+		]);
+
+		$response = curl_exec($curl);
+		$error = curl_error($curl);
+
+		curl_close($curl);
+		if (!empty($response)) {
+			$json_array = json_decode($response, true);
+			foreach($json_array['voices'] as $row) {
+				$voice_id = $row['voice_id'];
+				$name = $row['name'];
+				$gender = $row['labels']['gender'] ?? '';
+				$accent = $row['labels']['accent'] ?? '';
+				$use_case = $row['labels']['use case'] ?? '';
+				$recommended_model = $row['high_quality_base_model_ids'][0] ?? '';
+				$return_value[$voice_id] = "$name ($gender, $accent";
+				if (!empty($use_case)) {
+					$return_value[$voice_id] .= ", " . $use_case;
+				}
+				$return_value[$voice_id] .= ")";
+				if (!empty($recommended_model)) {
+					$return_value[$voice_id] .= " - $recommended_model";
+				}
+			}
+		}
+		return $return_value;
+	}
+
+	public function set_language(string $audio_language) {
+		$this->languages = $audio_language;
+	}
+
+	public function set_model(string $model): void {
+		if (array_key_exists($model, $this->get_models())) {
+			$this->model = $model;
+		}
+	}
+
+	public function get_model() {
+
+			//if the voice is not set return the default model
+			if (empty($this->voice)) {
+				return 'eleven_monolingual_v1';
+			}
+
+			//get the voices and automatically find the model
+			$url = 'https://api.elevenlabs.io/v1/voices';
+			$headers = [
+				'Content-Type: application/json',
+				"xi-api-key: $this->api_key",
+			];
+			$curl = curl_init();
+			curl_setopt_array($curl, [
+				CURLOPT_URL => $url,
+				CURLOPT_RETURNTRANSFER => true,
+				CURLOPT_ENCODING => "",
+				CURLOPT_MAXREDIRS => 10,
+				CURLOPT_TIMEOUT => 30,
+				CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
+				CURLOPT_CUSTOMREQUEST => "GET",
+			]);
+			$response = curl_exec($curl);
+			$error = curl_error($curl);
+			curl_close($curl);
+			if (!empty($response)) {
+				$json_array = json_decode($response, true);
+				foreach($json_array['voices'] as $row) {
+					if ($this->voice == $row['voice_id'] && !empty($row['high_quality_base_model_ids'][0])) {
+						return $row['high_quality_base_model_ids'][0];
+					}
+				}
+				return 'eleven_monolingual_v1';
+			}
+	}
+
+	public function get_models(): array {
+		return [
+			'eleven_monolingual_v1' => 'Default',
+			'eleven_turbo_v1' => 'Eleven Turbo v1',
+			'eleven_turbo_v2' => 'Eleven Turbo v2',
+			'eleven_multilingual_v1' => 'Eleven Multilingual v1',
+			'eleven_multilingual_v2' => 'Eleven Multilingual v2',
+		];
+	}
+}
+
+?>

+ 219 - 0
resources/classes/speech_openai.php

@@ -0,0 +1,219 @@
+<?php
+
+
+ /**
+ * ai class
+ *
+ * @method null download
+ */
+if (!class_exists('ai_openai')) {
+	class speech_openai implements speech_interface {
+
+		/**
+		 * declare private variables
+		 */
+		private $api_key;
+		private $path;
+		private $filename;
+		private $format;
+		private $voice;
+		private $message;
+		private $model;
+
+		/**
+		 * called when the object is created
+		 */
+		public function __construct($settings) {
+
+			//build the setting object and get the recording path
+			$this->api_key = $settings->get('speech', 'api_key');
+
+		}
+
+		public function set_path(string $audio_path) {
+			$this->path = $audio_path;
+		}
+
+		public function set_filename(string $audio_filename) {
+			$this->filename = $audio_filename;
+		}
+
+		public function set_format(string $audio_format) {
+			$this->format = $audio_format;
+		}
+
+		public function set_voice(string $audio_voice) {
+			$this->voice = $audio_voice;
+		}
+
+		public function set_language(string $audio_language) {
+			$this->language = $audio_language;
+		}
+
+		public function set_translate(string $audio_translate) {
+			$this->translate = $audio_translate;
+		}
+
+		public function set_message(string $audio_message) {
+			$this->message = $audio_message;
+		}
+
+		public function is_language_enabled() : bool {
+			//return the whether engine is handles languages
+			return false;
+		}
+
+		public function is_translate_enabled() : bool {
+			//return the whether engine is able to translate
+			return false;
+		}
+
+		public function get_voices() : array {
+			$voices = array(
+				"alloy" => "alloy",
+				"echo" => "echo",
+				"fable" => "fable",
+				"nova" => "nova",
+				"onyx" => "onyx",
+				"shimmer" => "shimmer"
+			);
+
+			//return the languages array
+			return $voices;
+		}
+
+		public function get_languages() : array {
+			//create the languages array
+			$languages = array(
+				"af" => "Afrikaans",
+				"ar" => "Arabic",
+				"hy" => "Armenian",
+				"az" => "Azerbaijani",
+				"be" => "Belarusian",
+				"bs" => "Bosnian",
+				"bg" => "Bulgarian",
+				"ca" => "Catalan",
+				"zh" => "Chinese",
+				"hr" => "Croatian",
+				"cs" => "Czech",
+				"da" => "Danish",
+				"nl" => "Dutch",
+				"en" => "English",
+				"et" => "Estonian",
+				"fi" => "Finnish",
+				"fr" => "French",
+				"gl" => "Galician",
+				"de" => "German",
+				"el" => "Greek",
+				"he" => "Hebrew",
+				"hi" => "Hindi",
+				"hu" => "Hungarian",
+				"is" => "Icelandic",
+				"id" => "Indonesian",
+				"it" => "Italian",
+				"ja" => "Japanese",
+				"kn" => "Kannada",
+				"kk" => "Kazakh",
+				"ko" => "Korean",
+				"lv" => "Latvian",
+				"lt" => "Lithuanian",
+				"mk" => "Macedonian",
+				"ms" => "Malay",
+				"mr" => "Marathi",
+				"mi" => "Maori",
+				"ne" => "Nepali",
+				"no" => "Norwegian",
+				"fa" => "Persian",
+				"pl" => "Polish",
+				"pt" => "Portuguese",
+				"ro" => "Romanian",
+				"ru" => "Russian",
+				"sr" => "Serbian",
+				"sk" => "Slovak",
+				"sl" => "Slovenian",
+				"es" => "Spanish",
+				"sw" => "Swahili",
+				"sv" => "Swedish",
+				"tl" => "Tagalog",
+				"ta" => "Tamil",
+				"th" => "Thai",
+				"tr" => "Turkish",
+				"uk" => "Ukrainian",
+				"ur" => "Urdu",
+				"vi" => "Vietnamese",
+				"cy" => "Welsh"
+			);
+
+			//return the languages array
+			return $languages;
+		}
+
+		/**
+		 * speech - text to speech
+		 */
+		public function speech() : bool {
+
+			// set the request URL
+			$url = 'https://api.openai.com/v1/audio/speech';
+
+			// set the request headers
+			$headers = [
+				'Authorization: Bearer ' . $this->api_key,
+				'Content-Type: application/json'
+			];
+
+			// set the http data
+			$data['model'] = 'tts-1-hd';
+			$data['input'] = $this->message;
+			$data['voice'] = $this->voice;
+			$data['response_format'] = 'wav';
+			if (isset($this->language)) {
+				$data['language'] = $this->language;
+			}
+			if (isset($this->translate)) {
+				$data['task'] = 'translate';
+			}
+
+			// initialize curl handle
+			$ch = curl_init($url);
+
+			// set the curl options
+			curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+			curl_setopt($ch, CURLOPT_CUSTOMREQUEST, 'POST');
+			curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
+			curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($data));
+
+			// run the curl request and get the response
+			$response = curl_exec($ch);
+
+			// close the handle
+			curl_close($ch);
+
+			// check for errors
+			if ($response === false) {
+				return false;
+			}
+			else {
+				// save the audio file
+				file_put_contents($this->path.'/'.$this->filename, $response);
+				return true;
+			}
+
+		}
+
+		public function set_model(string $model): void {
+			if (array_key_exists($model, $this->get_models())) {
+				$this->model = $model;
+			}
+		}
+
+		public function get_models(): array {
+			return [
+				'tts-1-hd' => 'tts-1-hd'
+			];
+		}
+
+	}
+}
+
+?>

+ 21 - 0
resources/interfaces/speech_interface.php

@@ -0,0 +1,21 @@
+<?php
+
+//define the template class
+if (!interface_exists('speech_interface')) {
+	interface speech_interface {
+		public function get_languages() : array;
+		public function get_models(): array;
+		public function get_voices() : array;
+		public function is_language_enabled() : bool;
+		public function set_filename(string $audio_filename);
+		public function set_format(string $audio_format);
+		public function set_language(string $audio_language);
+		public function set_message(string $audio_message);
+		public function set_model(string $audio_model): void;
+		public function set_path(string $audio_path);
+		public function set_voice(string $audio_voice);
+		public function speech() : bool;
+	}
+}
+
+?>