Instruction API en français

16-05-2023

L'API n'est disponible qu'une fois que vous avez approvisionné votre compte..

Voix de l'API JSON https://speechgen.io/index.php?r=api/voices.

Deux options de voix-off textuelle sont disponibles via l'API :

1 options - voix off rapide de textes courts.
2 option - définir une tâche de voix-off.

1 option - voix off rapide de textes courts

Url pour la variante 1: https://speechgen.io/index.php?r=api/text

Le texte est envoyé à l'API, vous obtenez des résultats immédiats..

Cette option a une limite maximale de 2000 caractères. Et aussi un maximum de 2 changements de voix. Si les limites sont dépassées, vous obtiendrez une erreur (statut = -1)..

Format de la demande

$data = [
	'token'=>'SekretKeyInYourProfile',
	'email'=>'yourMail@gmail.com',
	'voice'=>'John',
	'text'=>"Text for text to Speech", 
	'format'=>'mp3',
	'speed'=>1.1, 
	'pitch'=>0.8,   
	'emotion'=>'good', 
	];

Paramètres obligatoires :

'token' - 'SekretKeyInYourProfile'
'email' - 'yourMail@gmail.com'
'voice' - 'Voice'
'text' - 'Text for text to Speech' 
 

Paramètres facultatifs :

'format' - format of the resulting file, by default = mp3, possible values ( 'mp3', 'wav', 'ogg')
'speed' - playback speed, by default 1, (range from  0.1 to 2.0) 
'pitch'- voice pitch, by default 0, (range from -20 to 20)    
'emotion' - emotion of voice , by default 'good', possible values ( 'good',  'evil', 'neutral'). 
Not available for all voices,  look it here.

La réponse est en JSON :

{
"id":"4153594",
"status":"0",
"parts":"5",
"parts_done":"2",
"duration":"0",
"format":"mp3",
"error":"",
"balans":"3331.2720000314",
"cost":"0.06"
}

Description :

"id" - unique voice ID
"status" - current voiceover status. Available from 3 values:
    0  - process
    1  - completed successfully
    -1 - error 
"file" - audio file path, available if status= 1
"error" - error text, in the event of, if status =  -1
"parts"  - number of voiceovers
"parts_done" - number of pieces completed
"duration"  - audio file duration in seconds, available if status  = 1
"format" - audio file format
"balans" - limit balance
"cost" - voiceover costs. (is increased as the voice parts are dubbed)


 

<?
$data = [
    'token' => '123456',
    'email' => 'mail@mail.com',
    'voice' => 'John',
    'text' => "Text",
    'format' => 'mp3',
    'speed' => 1.1,
    'pitch' => 0,
    'emotion' => 'good',
];
$url = "https://speechgen.io/index.php?r=api/text";
 
$ch = curl_init();
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, false);
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, http_build_query($data));
 
$response = curl_exec($ch);
var_dump($response);
$response = json_decode($response, true);
if (curl_errno($ch)) {
    var_dump("Connection error with text recognition server, " . curl_error($ch));
} else {
    var_dump($response);
    if ($response["status"] == 1) {
        //Copy
        echo " ok " . $response["file"];
        copy($response["file"], 'Filename.' . $response["format"]);
    } else {
        //Error, no voiceover possible
        echo $response["error"];
    }
}
curl_close($ch);
?>
 

L'option 2 consiste à définir une tâche de voix off pour le texte..

Url pour l'option 2 : https://speechgen.io/index.php?r=api/longtext

Étape 1 : Le texte est envoyé à l'API et un identifiant vocal unique est obtenu (id).
Étape 2:  Demander le résultat du doublage en envoyant l'identifiant du doublage (id).

Cette méthode n'a aucune limite et permet de doubler des textes jusqu'à 1 000 000 de caractères. Le doublage s'effectue par ordre de priorité (premier arrivé, premier servi) et peut prendre de 1 à plusieurs minutes, en fonction de la longueur du texte.

Format de la demande

$data = [
	'token'=>'SekretKeyInYourProfile',
	'email'=>'yourMail@gmail.com',
	'voice'=>'John',
	'text'=>"Text", 
	'format'=>'mp3',
	'speed'=>1.1, 
	'pitch'=>0.8,   
	'emotion'=>'good', 
	];

Paramètres obligatoires:

'token' - 'SekretKeyInYourProfile'
'email' - 'yourMail@gmail.com'
'voice' - 'Voice'
'text' - 'Text' 
 

Paramètres facultatifs:

'format' - format of the resulting file, by default = mp3, possible values( 'mp3', 'wav', 'ogg')
'speed' - speed, by default 1, (range from 0.1 to 2.0) 
'pitch'- voice pitch, by default 0, (range from -20 to 20)    
'emotion' - emotional coloration, by default  'good', possible values ( 'good',  'evil', 'neutral'). 
Not available for all voices, look here.

La réponse est fournie en JSON :

{
"id":"4153594",
"status":"0",
"parts":"5",
"parts_done":"0",
"format":"mp3",
"error":"",
"balans":"3331.2720000314",
"cost":"0.00"
}

Description :

"id" - unique voice ID
"status" - current voiceover status. Available from 2 values:
    1  - task added successfully
    -1 - error 
"error" - error text, in case of, if status =  -1
"parts"  - number of voiceovers
"parts_done" - number of pieces completed
"format" - audio file format
"balans" - the balance of the account limits
"cost" - voiceover costs. ( is increased as the voice parts are dubbed)
 

Step 2. Après la création d'une tâche et l'obtention d'un identifiant (id), une requête doit être envoyée à l'adresse suivante https://speechgen.io/index.php?r=api/result, pour connaître le résultat.

Format de la demande

$data = [
	'token'=>'SekretKeyInYourProfile',
	'email'=>'yourMail@gmail.com',
	'id'=>4153594, 
];

Paramètres obligatoires:

'token' - SekretKeyInYourProfile
'email' - yourMail@gmail.com
'id' - Unique voice ID obtained in step 1
 

La réponse est fournie en JSON:

{
"id":"4153594",
"status":"0",
"file":"result.mp3",
"cuts":["result_1.mp3","result_2.mp3",...],
"parts":"5",
"parts_done":"0",
"format":"mp3",
"error":"",
"balans":"3331.2720000314",
"cost":"0.00"
}

Description :

"id" - unique voice ID
"status" - current voiceover status. Available from 3 values:
    0  - in the midst of
    1  - completed successfully
    -1 - error 
"file" - audio file path, available if status = 1
"error" - error text, in case of, if status =  -1
"parts"  - number of voiceovers
"parts_done" - number of pieces completed
"duration"  - audio file duration in seconds, available if status = 1
"format" - audio file format
"balans" - the balance of the account limits
"cost" - voiceover costs. ( is increased as the voice parts are dubbed)
"cuts" - Array of fragments, if the "cut" tag is used in the text

 

<?
//STEP 1
$data = [
    'token' => '123456',
    'email' => 'mail@mail.com',
    'voice' => 'John',
    'text' => "Text",
    'format' => 'mp3',
    'speed' => 1.1,
    'pitch' => 0,
    'emotion' => 'good',
];
$url = "https://speechgen.io/index.php?r=api/longtext";
 
$ch = curl_init();
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, false);
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, http_build_query($data));
 
$response = curl_exec($ch);
var_dump($response);
$response = json_decode($response, true);
$resultId = 0;
if (curl_errno($ch)) {
    var_dump("Error in connecting to the text recognition server, " . curl_error($ch));
} else {
    var_dump($response);
    if ($response["status"] == 1) {
        //Copy
        echo " ok " . $response["file"];
        copy($response["file"], 'FileName.' . $response["format"]);
    } elseif ($response["status"] == 0) {
        //Remember the voice ID, and ask for the result later
        $resultId = $response["id"];
    } else {
        //Error, no voiceover possible
        echo $response["error"];
    }
}
curl_close($ch);
 
 
 
//STEP 2
// Every minute we make a query-check of the result
if($resultId){
    $data = [
        'token'=>'123456',
        'email'=>'mail@mail.com',
        'id'=>$resultId, 
    ];    
    $url = "https://speechgen.io/index.php?r=api/result";
 
    $ch = curl_init();
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($ch, CURLOPT_URL, $url);
    curl_setopt($ch, CURLOPT_HEADER, false);
    curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, false);                
    curl_setopt($ch, CURLOPT_POST, 1);
    curl_setopt($ch, CURLOPT_POSTFIELDS, http_build_query($data));    
 
    $response = curl_exec($ch); 
    var_dump($response);
    $response = json_decode($response,true);
    if (curl_errno($ch)) {
        var_dump("Error in connecting to the text recognition server, ".curl_error($ch));
    }else{ 
        var_dump($response);
        if($response["status"] == 1){
            //Copy
            echo "  ok ".$response["file"];
            copy($response["file"], 'FileMane.' . $response["format"]);
        }elseif($response["status"] == 0){
            //Not ready yet, we will ask for the result later
            echo "In process...";
        }else{
            //Error, no voiceover possible
            echo $response["error"];
        }
    }
    curl_close($ch);
}
 
 
?>