Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/phpaisdk/elevenlabs/llms.txt

Use this file to discover all available pages before exploring further.

The Voice Changer performs speech-to-speech conversion: it takes any input audio recording and transforms it into a chosen ElevenLabs voice while preserving the original prosody, rhythm, and emotional delivery. This makes it ideal for dubbing performances, re-voicing podcast episodes, or applying a branded voice to recorded content without re-recording from scratch.

​
Usage

Call ElevenLabs::voiceChanger()->convert() with a target voice ID, a Content::audio() value wrapping the source file, and an optional options array. 
use AiSdk\Content;
use AiSdk\ElevenLabs;

$result = ElevenLabs::voiceChanger()->convert(
    voiceId: 'JBFqnCBsd6RMkjVDRZzb',
    audio: Content::audio(__DIR__.'/performance.wav'),
    options: [
        'model_id' => 'eleven_multilingual_sts_v2',
        'remove_background_noise' => true,
        'output_format' => 'wav_44100',
    ],
);

$result->output->save(__DIR__.'/changed.wav');
You can also call the same method on a configured provider instance rather than the static facade: 
use AiSdk\Content;
use AiSdk\ElevenLabs;

$elevenLabs = ElevenLabs::create(['apiKey' => 'xi-...']);

$result = $elevenLabs->voiceChanger()->convert(
    voiceId: 'JBFqnCBsd6RMkjVDRZzb',
    audio: Content::audio(__DIR__.'/performance.wav'),
    options: [
        'model_id' => 'eleven_multilingual_sts_v2',
        'remove_background_noise' => true,
        'output_format' => 'wav_44100',
    ],
);

$result->output->save(__DIR__.'/changed.wav');

​
Parameters

​
voiceId
string
required
The ElevenLabs voice ID to convert the source audio into. Any voice ID from your ElevenLabs workspace is accepted, for example JBFqnCBsd6RMkjVDRZzb.
​
audio
Content
required
The source audio to transform, wrapped in a Content::audio() value. Accepts a file path or inline bytes:
// From a file path
Content::audio(__DIR__.'/performance.wav')

// From raw bytes with explicit MIME type and filename
Content::audio($rawBytes, 'audio/wav', 'performance.wav')
​
options
array
An optional associative array of additional request fields. Most fields are sent as multipart form parts; a small set of fields are promoted to query parameters automatically — see Query parameters below.

​
Options

​
Request body fields

These options are sent in the multipart request body alongside the audio file.
​
model_id
string
The speech-to-speech model to use. ElevenLabs recommends eleven_multilingual_sts_v2 for high-quality multilingual conversion.
​
remove_background_noise
boolean
When true, background noise is stripped from the input audio before conversion begins. Useful when the source recording was captured in a noisy environment.
​
voice_settings
object
Fine-grained voice control passed as a JSON object. Accepts stability, similarity_boost, style, and use_speaker_boost keys as documented by ElevenLabs.

​
Query parameters

The following fields are automatically extracted from the $options array and appended to the request URL as query parameters instead of being sent in the body. This matches the ElevenLabs API contract for the speech-to-speech endpoint.
​
output_format
string
The desired audio encoding and sample rate of the returned file. Common values include mp3_44100_128, wav_44100, and pcm_16000. Defaults to mp3_44100_128 when omitted.
​
enable_logging
boolean
When false, the request and response are excluded from ElevenLabs usage logs and history.
​
optimize_streaming_latency
integer
A legacy latency-optimisation level (0–4). Deprecated by ElevenLabs but still accepted. Omit this field unless you have an existing integration that depends on it.
output_format, enable_logging, and optimize_streaming_latency are silently removed from the multipart body and sent as query parameters. All other keys in $options remain in the request body.

​
Return value

convert() returns an AudioResult. Access the converted audio through its output property:
​
output
AudioData
The converted audio file.
Save directly to a file path using the helper method: 
$result->output->save(__DIR__.'/changed.wav');
Or work with the raw bytes directly: 
$bytes = $result->output->data;
file_put_contents('/tmp/changed.wav', $bytes);

​
Full example

use AiSdk\Content;
use AiSdk\ElevenLabs;

ElevenLabs::create(['apiKey' => $_ENV['ELEVENLABS_API_KEY']]);

$result = ElevenLabs::voiceChanger()->convert(
    voiceId: 'JBFqnCBsd6RMkjVDRZzb',
    audio: Content::audio(__DIR__.'/performance.wav'),
    options: [
        'model_id'               => 'eleven_multilingual_sts_v2',
        'remove_background_noise' => true,
        'output_format'          => 'wav_44100',
        'enable_logging'         => false,
    ],
);

$result->output->save(__DIR__.'/changed.wav');

echo 'Saved converted audio — ' . strlen($result->output->data) . ' bytes';

Build docs developers (and LLMs) love

Get started for freeTalk to us