Lara's Translate Audio feature enables you to automate the localisation of audio content through a direct audio-to-audio workflow. The system analyses the source track and generates a new translated version.
Thanks to native integration with Lara's computer-assisted translation ecosystem, the end result is not a simple, literal translation, but an output optimised for your specific linguistic domain and communication style.

Key Features:

Adaptation to Context (Translation Memories & Glossaries): Through the ability to use translation memories and glossaries, the API ensures that technical terms, product names, and preferred terminology are translated consistently throughout the audio, in line with the linguistic resources available in your account.
Control over Translation Style: Lara lets you shape the translated audio through the style parameter. Choose Faithful for a precise translation that preserves the original structure and meaning, Fluid for a smoother and more natural result, or Creative for a more expressive adaptation suited to marketing, literary, or highly engaging content.
Voice Gender Selection: Through the voiceGender parameter, the system allows you to specify the vocal characteristics of the output. You can choose between male or female voices to ensure the translated audio perfectly aligns with your brand identity or the nature of the original speaker.
Automatic Language Detection: The system can independently identify the language spoken in the source audio file, simplifying integration into platforms that manage content from international users.
Extensive Multilingual Support: The API supports a wide range of language codes (from the most common ones, such as en-US and it-IT, to less widely used languages), enabling immediate global scalability of your audio assets.

Translate

To begin, you must first instantiate the Translator class. This is the mandatory first step for all operations within the Lara SDK.

Initialization Guide →

All audio translation functionalities can be accessed via the same Translator (lara in the example) object of the above example, through the AudioTranslator class.

Audio translation is an asynchronous process consisting of three phases.

Upload the audio file
Wait for Lara to finish the process, polling the status at regular intervals until it changes to translated
Download the file

Once uploaded with the upload method, Lara will return a Audio object; the id of this object can be used both to check the status with the status method, both to download the translated file with the download method once its status became translated.

Audio object

Field	Type	Description
id	String	Unique ID of the document. Format: doc_xyz123
status	String	The current status of the translation.
source	String	The source language of the document.
target	String	The target language of the translation.
voiceGender	String	Gender of the voice in the translation, available values: `male`, `female`
filename	String	Filename of the source file.
createdAt	DateTime	When the document is created.
updatedAt	DateTime	When the document is last updated (equal to created_at with no updates).
options	AudioOptions	Options related to the document translation.
translatedSeconds	Integer	The number of translated seconds.
totalSeconds	Integer	The number of seconds in the document.
errorReason	String	The reason of the error, if any.

Translate method

The SDK contains a translate method that simplifies the process of translating audio file by uploading the file, checking the status at regular intervals, and returning the result of the download function.

Request

sample_audio_file = "sample_audio.mp3"

translated_audio = lara.audio.translate(
				file_path=sample_audio_file,
				filename=os.path.basename(sample_audio_file),
				source="en-US",
				voice_gender = "female",
				target="it-IT",
				adapt_to=["mem_1234..."],
				glossaries=["gls_1234..."]
        )

const options: AudioTranslateOptions = {
  adaptTo: ['mem_1_id', 'mem_2_id'],
  glossaries: ['gls_1_id', 'gls_2_id'],
	voiceGender : 'male',
  style: 'fluid'
};

// Translate an audio file
const translation = await lara.audio.translate(file, 'test.wav', 'en', 'it', options);

File audioFile = new File("voice.mp3");

AudioUploadOptions options = new AudioUploadOptions();
options.setAdaptTo("mem_1234...");
options.setGlossaries("gls_1234...");
options.setVoiceGender = "female";

InputStream translated = lara.audio.translate(audioFile, "en-US", "it-IT", options);

$translateOptions = new AudioTranslateOptions([
  "adaptTo" => ["mem_1_id", "mem_2_id"],
  "glossaries" => ["gls_1_id", "gls_2_id"],
  "voiceGender" => "male",
  "style" => "fluid"
]);

// Translate an audio file
$translation = $lara->audio->translate($file_path, "en", "it", $translateOptions);

// Sdk available soon

val audioFile = File("voice.mp3")
val options = AudioUploadOptions().apply {
    adaptTo = "mem_1234..."
  	glossaries = "gls_1234..."
		voiceGender = "female"

}
val translated: InputStream = lara.audio.translate(audioFile, "en-US", "it-IT", options)

var options = new AudioTranslateOptions {
    AdaptTo = new[] { "mem_1_id", "mem_2_id" },
    Glossaries = new[] { "gls_1_id", "gls_2_id" },
    Style = TranslationStyle.Fluid
};

// Translate an audio file
var translation = await lara.Audio.Translate("path/to/file", "en", "it", options);

let sampleFilePath = "sample_audio.mp3"  // Create this file with your content

let sourceLang = "en-US"
let targetLang = "de-DE"
let gender = "female"

print("Translating audio: \(FileManager.default.displayName(atPath: sampleFilePath)) from \(sourceLang) to \(targetLang)")

        let audioData = try Data(contentsOf: URL(fileURLWithPath: sampleFilePath))
        let translatedData = try await lara.audio.translate(
	          data: audioData,
						voiceGender = gender,
            filename: FileManager.default.displayName(atPath: sampleFilePath),
            source: sourceLang,
            target: targetLang
        )

  // Save translated audio - replace with your desired output path
  let outputPath = "sample_audio_translated.mp3"
  try translatedData.write(to: URL(fileURLWithPath: outputPath))

Here follows the basic fields for the translate method:

Field	Type	Required	Default	Description
file / filePath	File / String	Yes		The input file/the path to input file to translate.
filename	String	Yes*		Only the Typescript/Javascript SDK requires the filename as parameter.
source	String	No	Autodetected	The source language code (e.g., "en-EN" for English). If not specified, the system will attempt to detect it automatically.
target	String	Yes		The target language code (e.g., "it-IT" for Italian). This specifies the language you want the audio translated into.
options	AudioTranslateOptions	No		See the table below for details.

Some options are available to customize the behavior of the translate method:

Field	Type	Required	Default	Description
adaptTo	String[]	No	Default is all Memories on your account	A list of translation memory IDs for adapting the translation.
glossaries	String[]	No		A list of glossary IDs.
noTrace	Boolean	No	False	If set to True, source content and its translation will not be saved on our system. (AKA Incognito mode)
style	String	No	`faithful`	The style to apply to the translation. Available values: `faithful` `fluid` `creative`

The following sections provide a detailed breakdown of the phases and methods utilized by the translate method.

Upload an audio file

The upload method will take care of uploading the source file to Lara and starting the translation process.

Request

sample_audio_file = "sample_audio.mp3"

audio = lara.audio.upload(
            file_path=sample_audio_file,
            filename=os.path.basename(sample_audio_file),
            source="en-US",
            target="it-IT",
            adapt_to=["mem_1234..."]
  					glossaries=["gls_1234..."]
)

const options: AudioUploadOptions = {
  adaptTo: ['mem_1_id', 'mem_2_id'],
  glossaries: ['gls_1_id', 'gls_2_id'],
  style: 'fluid',
};

// Upload an audio file and start the translation process
const audio: Audio = await lara.audio.upload(file, 'test.wav', 'en', 'it', options);

File audioFile = new File("voice.mp3");

AudioUploadOptions options = new AudioUploadOptions();
options.setAdaptTo("mem_1234...");
options.setGlossaries("gls_1234...");

Audio audio = lara.audio.upload(audioFile, "en-US", "it-IT", options);

$translateOptions = new AudioUploadOptions([
  "adaptTo" => ["mem_1_id", "mem_2_id"],
  "glossaries" => ["gls_1_id", "gls_2_id"],
  "style" => "fluid"
]);

// Upload a audio and start the translation process
$audio = $lara->audio->upload($file_path, "en", "it", $translateOptions);

// Sdk available soon

val audioFile = File("voice.mp3")
val options = AudioUploadOptions().apply {
    adaptTo = "mem_1234..."
    glossaries = "gls_1234..."
}
val audio: Audio = lara.audio.upload(audioFile, "en-US", "it-IT", options)

var options = new AudioUploadOptions {
    AdaptTo = new[] { "mem_1_id", "mem_2_id" },
    Glossaries = new[] { "gls_1_id", "gls_2_id" },
    Style = TranslationStyle.Fluid
  };

var audio = await lara.Audio.Upload("path/to/file", "en", "it", options);

 let sampleFilePath = "sample_audio.mp3"  // Create this file with your content

 let sourceLang = "en-US"
 let targetLang = "de-DE"

 let audioData = try Data(contentsOf: URL(fileURLWithPath: sampleFilePath))
 let audio = try await lara.audio.upload(
          data: audioData,
          filename: FileManager.default.displayName(atPath: sampleFilePath),
          source: sourceLang,
          target: targetLang,
          options: AudioUploadOptions(
              adaptTo: ["mem_1A2b3C4d5E6f7G8h9I0jKl"],  // Replace with actual memory IDs
              glossaries: ["gls_1A2b3C4d5E6f7G8h9I0jKl"]  // Replace with actual glossary IDs
            )
 )
// Check status with polling
var updatedAudio = try await lara.audio.status(id: audio.id)
while updatedAudio.status != .translated {
updatedAudio = try await lara.audio.status(id: audio.id)
	if updatedAudio.status == .error {
  	throw NSError(domain: "AudioTranslationError",
    	code: 500,
      userInfo: [NSLocalizedDescriptionKey: updatedAudio.errorReason ?? "Translation failed"])
      }
       try await Task.sleep(nanoseconds: 2_000_000_000) // 2 seconds
    }

        // Download translated audio
        let translatedData = try await lara.audio.download(id: audio.id)

        // Save translated audio - replace with your desired output path
        let outputPath = "step_audio_translated.mp3"
try translatedData.write(to: URL(fileURLWithPath: outputPath))

Here follows the basic fields for the upload method:

Field	Type	Required	Default	Description
file	File	Yes		The input file to translate.
filename	String	Yes*		The Typescript/Javascript and Python SDKs require the filename as a parameter.
source	String	No	Autodetected	The source language code (e.g., "en-EN" for English). If not specified, the system will attempt to detect it automatically.
target	String	Yes		The target language code (e.g., "it-IT" for Italian). This specifies the language you want the audio translated into.
options	AudioUploadOptions	No		See the table below for details.

There are some options available to customize the behavior of the upload method:

Field	Type	Required	Default	Description
adaptTo	String[]	No	Default is all Memories on your account	A list of translation memory IDs for adapting the translation.
glossaries	String[]	No		A list of glossary IDs.
noTrace	Boolean	No	False	If set to True, source content and its translation will not be saved on our system.
style	String	No	`faithful`	The style to apply to the translation. Available values: `faithful` `fluid` `creative`

The result of the upload method is a Audio object that is needed to check the translation status and download the translated audio file.

Response

# audio
audio = Audio(
    status="initialized",
    id="doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename="voice.mp3",
    updatedAt="2025-05-29T18:00:24.779000+00:00",
    createdAt="2025-05-29T18:00:24.779000+00:00",
    source="en",
    target="it"
)

// audio
{
  status: 'initialized',
  id: 'doc_3TJ2ObqnrYTC2k5vSfLlqz',
  filename: 'test.wav',
  updatedAt: 2025-05-15T09:28:14.626Z,
  createdAt: 2025-05-15T09:28:14.626Z,
  source: 'en',
  target: 'it'
}

// audio
Audio(
    status="initialized",
    id="doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename="voice.mp3",
    updatedAt=2025-05-15T09:28:14.626Z,
    createdAt=2025-05-15T09:28:14.626Z,
    source="en",
    target="it"
)

// audio
$audio = [
    "status" => "initialized",
    "id" => "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    "filename" => "test.wav",
    "updated_at" => "2025-05-29T18:00:24.779000+00:00",
    "created_at" => "2025-05-29T18:00:24.779000+00:00",
    "source" => "en",
    "target" => "it"
];

// Sdk available soon

val audio = Audio(
    status = "initialized",
    id = "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename = "voice.mp3",
    updatedAt = ZonedDateTime.parse("2025-05-15T09:28:14.626Z"),
    createdAt = ZonedDateTime.parse("2025-05-15T09:28:14.626Z"),
    source = "en",
    target = "it"
)

// audio
var audio = new Audio(
    id: "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    status: AudioStatus.Initialized,
    filename: "test.wav",
    target: "it",
    createdAt: "2025-05-15T09:28:14.626Z",
    updatedAt: "2025-05-15T09:28:14.626Z",
    source: "en"
);

// document
let document = Document(
    id: "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    status: .initialized,
    source: "en",
    target: "it",
    filename: "test.mp3",
    createdAt: "2025-05-15T09:28:14.626Z",
    updatedAt: "2025-05-15T09:28:14.626Z"
)

Check status

After the upload is finished, it is possible to check the status of the translation at regular intervals. This can be done with the status method:

Request

// Check the status of the audio translation
audio = lara.audio.status(audio.id)

// check the status of the audio file
audio = await lara.audio.status(audio.id);

// Check the status of the audio translation
audio = lara.audio.status(audio.getId());

// check the status of the audio file
$audio = $lara->audio->status($audio->getId());

// Sdk available soon

// Check the status of the audio translation
audio = lara.audio.status(audio.id)

// Check the status of the audio file
var audio = await lara.Audio.Status(audio.Id);

// Check the status of the audio file
$updatedAudio = $lara->audio->status($audio->getId());
echo "Current status: " . $updatedAudio->getStatus() . "\n";

The method takes as unique parameter the id of the Audio object obtained with the upload method.

Field	Type	Required	Default	Description
id	String	Yes		The audio file id obtained with the `upload` method.

Response

# audio
audio = Audio(
    status="initialized",
    id="doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename="vioce.mp3",
    updatedAt="2025-05-29T18:00:24.779000+00:00",
    createdAt="2025-05-29T18:00:24.779000+00:00",
    source="en",
    target="it"
)

// document
{
  status: 'translated',
  id: 'doc_3TJ2ObqnrYTC2k5vSfLlqz',
  filename: 'test.mp3',
  updatedAt: 2025-05-15T09:28:22.347Z,
  createdAt: 2025-05-15T09:28:14.626Z,
  source: 'en',
  target: 'it'
}

// audio
Audio(
    status="initialized",
    id="doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename="voice.mp3",
    updatedAt=2025-05-15T09:28:14.626Z,
    createdAt=2025-05-15T09:28:14.626Z,
    source="en",
    target="it"
)

// audio
$audio = [
    "status" => "translated",
    "id" => "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    "filename" => "test.mp3",
    "updated_at" => "2025-05-29T18:00:32.519000+00:00",
    "created_at" => "2025-05-15T09:28:14.626",
    "source" => "en",
    "target" => "it"
];

// Sdk available soon

// Sdk available soon
val audio = Audio(
    status = "initialized",
    id = "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    filename = "voice.mp3",
    updatedAt = ZonedDateTime.parse("2025-05-15T09:28:14.626Z"),
    createdAt = ZonedDateTime.parse("2025-05-15T09:28:14.626Z"),
    source = "en",
    target = "it"
)

// audio
var audio = new audio(
    id: "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    status: DocumentStatus.Translated,
    filename: "test.mp3",
    target: "it",
    createdAt: "2025-05-15T09:28:14.626Z",
    updatedAt: "2025-05-15T09:28:22.347Z",
    source: "en"
);

// document
let audio = audio(
    id: "doc_3TJ2ObqnrYTC2k5vSfLlqz",
    status: .initialized,
    source: "en",
    target: "it",
    filename: "test.mp3",
    createdAt: "2025-05-15T09:28:14.626Z",
    updatedAt: "2025-05-15T09:28:14.626Z"
)

Download the translation

After updating the audio file status with the status method and setting it to translated, you can download the translated audio file with the download method.

Request

// Download the translation
translated_audio = lara.audio.download(audio.id)

// Download the translation
const translation = await lara.audio.download(audio.id);

// Download the translation
InputStream downloaded = lara.audio.download(audio.getId())

// Download the translation
$translation = $lara->audio->download($audio->getId());

// Sdk available soon

// Download the translation
val downloaded: InputStream = lara.audio.download(audio.id)

// Download the translation
var translation = await lara.Audio.Download(audio.Id);

// Download the translation
let translation = try await lara.audio.download(id: document.id)

Here follows the basic fields for the download method:

Field	Type	Required	Default	Description
id	String	Yes		The id of the document obtained with the `upload` method.

Supported Languages

Afrikaans - af-ZA

Albanian - sq-AL

Amharic - am-ET

Arabic - ar-SA Armenian - hy-AM

Azerbaijani - az-AZ

Basque - eu-ES

Bengali - bn-BD Bulgarian - bg-BG

Burmese - my-MM

Catalan - ca-ES

Chinese (Simplified) - zh-CN

Chinese (Traditional, Hong Kong) - zh-HK

Croatian - hr-HR

Czech - cs-CZ

Danish - da-DK

Dutch - nl-NL

English (Australia) - en-AU

English (Canada) - en-CA

English (Ireland) - en-IE

English (United Kingdom) - en-GB

English (United States) - en-US

Estonian - et-EE

Filipino - fil-PH

Finnish - fi-FI

French - fr-FR

Galician - gl-ES

Georgian - ka-GE

German - de-DE

Hebrew - he-IL

Hindi - hi-IN

Hungarian - hu-HU

Icelandic - is-IS

Indonesian - id-ID

Irish - ga-IE

Italian - it-IT

Japanese - ja-JP

Javanese - jv-ID

Kazakh - kk-KZ

Korean - ko-KR

Lao - lo-LA

Latvian - lv-LV

Lithuanian - lt-LT

Macedonian - mk-MK

Malay - ms-MY

Malayalam - ml-IN

Maltese - mt-MT

Marathi - mr-IN

Mongolian - mn-MN

Norwegian Bokmål - nb-NO

Pashto - ps-PK Persian - fa-IR

Polish - pl-PL

Portuguese (Brazil) - pt-BR

Portuguese (Portugal) - pt-PT

Romanian - ro-RO

Russian - ru-RU

Serbian (Cyrillic script) - sr-Cyrl-RS Slovak - sk-SK

Slovenian - sl-SI

Somali - so-SO

Spanish - es-ES

Spanish (Argentina) - es-AR

Spanish (Latin America) - es-419 Spanish (Mexico) - es-MX

Swahili - sw-KE

Swedish - sv-SE

Tagalog - tl-PH Tamil - ta-IN

Thai - th-TH

Turkish - tr-TR

Ukrainian - uk-UA

Urdu - ur-PK Uzbek - uzn-UZ

Vietnamese - vi-VN

Welsh - cy-GB

Billing

More details are available in the pricing page