/**
* Sends audio file to API and a text representation is sent back.
*
* @method speechToText
*
* @param {string} file Name and path of local audio file to send to codekit
* @param {string} context Speech context for translation. Please see SpeechToText API documentation for parameter values
* @param {string} subcontext Speech subcontext for translation. Please see SpeechToText API documentation for parameter values
* @param {string} xargs X-Arg objects. Please see SpeechToText API documentation for information about this parameter
* @param {boolean} chunked True to send the file using chunked transfer.
* @param {string} language ISO language code (like 'en-US')
*
* @return {Response} Returns Response object.
* @throws ServiceException if API request was not successful.
*
*/
public function speechToText($file, $context, $subcontext, $xargs, $chunked, $language)
{
$filecontents = $this->getFile($file);
// throws Exception
// Get OAuth token
$token = $this->getCurrentClientToken();
$speechSrvc = new SpeechService($this->base_url, $token);
return $speechSrvc->speechToText($file, $context, $subcontext, $xargs, $chunked, $language, true);
}
public function testRequest()
{
require __DIR__ . '/cfgs/speech_config.php';
if (isset($proxyHost) && isset($proxyPort)) {
RestfulEnvironment::setProxy($proxyHost, $proxyPort);
}
if (isset($allowAllCerts)) {
RestfulEnvironment::setAcceptAllCerts($allowAllCerts);
}
$osrvc = new OAuthTokenService($FQDN, $api_key, $secret_key);
$token = $osrvc->getToken('Speech,TTS,STTC');
$srvc = new SpeechService($FQDN, $token);
$fname = __DIR__ . '/files/BostonCeltics.wav';
$response = $srvc->speechToText($fname, 'Generic');
$this->assertTrue($response != null);
$response = $srvc->textToSpeech('text/plain', 'testing ok');
$this->assertTrue($response != null);
$response = $srvc->speechToTextCustom('GenericHints', $fname);
$this->assertTrue($response != null);
}
public function handleSpeechToText()
{
if (!isset($_REQUEST['SpeechToText'])) {
return;
}
try {
$context = $_REQUEST['SpeechContext'];
$_SESSION['SpeechContext'] = $context;
$filename = $_REQUEST['audio_file'];
$_SESSION['audio_file'] = $filename;
$chunked = isset($_REQUEST['chkChunked']) ? $_REQUEST['chkChunked'] : false;
$_SESSION['chunked'] = $chunked;
$flocation = $this->_audioFolder . '/' . $filename;
$subContext = null;
if ($context == 'Gaming') {
$subContext = $this->_xSpeechSubContext;
}
$srvc = new SpeechService($this->apiFQDN, $this->getFileToken());
$response = $srvc->speechToText($flocation, $context, $subContext, $this->_xArgs, $chunked);
$this->results[SpeechController::RESULT_SPEECH_TO_TEXT] = $response;
} catch (Exception $e) {
$this->errors[SpeechController::ERROR_SPEECH_TO_TEXT] = $e->getMessage();
}
}
}
$speechSrvc = new SpeechService($FQDN, $token);
$speechContext = $_POST['speechContext'];
$speechFile = $_POST['speechFile'];
$xArg = $_POST['x_arg'];
$subContext = $_POST['x_subcontext'];
$chunked = isset($_POST['chunked']) ? true : null;
$allowedFiles = array('boston_celtics.wav', 'california.amr', 'coffee.amr', 'doctors.wav', 'nospeech.wav', 'samplerate_conflict_error.wav', 'this_is_a_test.spx', 'too_many_channels_error.wav', 'boston_celtics.wav');
if (!in_array($speechFile, $allowedFiles, true)) {
throw new Exception('Invalid speech file specified');
}
$flocation = $audioFolder . '/' . $speechFile;
if ($speechContext !== 'Gaming') {
$subContext = null;
}
$response = $speechSrvc->speechToText($flocation, $speechContext, $subContext, $xArg, $chunked);
$headers = null;
$values = null;
$nbest = $response->getNBest();
if ($nbest !== null) {
$headers = array('ResponseId', 'Status', 'Hypothesis', 'LanguageId', 'Confidence', 'Grade', 'ResultText', 'Words', 'WordScores');
$values = array($response->getResponseId(), $response->getStatus(), $nbest->getHypothesis(), $nbest->getLanguageId(), $nbest->getConfidence(), $nbest->getGrade(), $nbest->getResultText(), json_encode($nbest->getWords()), json_encode($nbest->getWordScores()));
} else {
$headers = array('ResponseId', 'Status');
$values = array($response->getResponseId(), $response->getStatus());
}
$arr = array('success' => true, 'tables' => array(array('caption' => 'Speech Response:', 'headers' => $headers, 'values' => array($values))));
} catch (Exception $e) {
$arr = array('success' => false, 'text' => $e->getMessage());
}
echo json_encode($arr);
// Get the OAuth access token.
$token = $osrvc->getToken('SPEECH,STTC,TTS');
// Create the service for interacting with the Speech API.
$speechSrvc = new SpeechService('https://api.att.com', $token);
// The Speech API requires the audio files to be certain formats. In order to
// convert speech files to the proper format, the ffmpeg program may be used.
// The ffmpeg program can be downloaded from https://ffmpeg.org/
// The following try/catch blocks can be used to test the methods of the
// Speech API. To test a specific method, comment out the other try/catch blocks.
/* This try/catch block tests the speechToText method. */
try {
// Enter the path of the file to translate. For example: $fname = '/tmp/file.wav';
$fname = 'ENTER VALUE!';
$speechContext = 'Generic';
// Send the request to convert the speech file to text.
$response = $speechSrvc->speechToText($fname, $speechContext);
echo 'responseId: ' . $response->getResponseId() . "\n";
} catch (ServiceException $se) {
echo $se->getErrorResponse();
}
/* This try/catch block tests the textToSpeech method. */
try {
// Specify the content type.
$ctype = 'ENTER VALUE!';
// Specify text to convert to speech.
$txt = 'ENTER VALUE!';
// Send the request to convert the specified text to audio.
$response = $speechSrvc->textToSpeech($ctype, $txt);
echo 'audio length: ' . strlen($response) . "\n";
} catch (ServiceException $se) {
echo $se->getErrorResponse();
