# SAM Audio Separate

AUDIOsam-audio-separate

**SAM Audio** (Segment Anything in Audio) from [Meta AI](https://ai.meta.com/) is a foundation model for general audio separation that unifies text, visual, and temporal span prompting in a single framework. Built on a diffusion transformer architecture trained with flow matching on large-scale audio data, it isolates specific sounds from a mixture based on a natural language description — returning both the isolated target and the residual. Supports speech, music, and general sound separation.

Best for

Isolating specific sounds from audio mixtures via text prompts, Stem separation for music and speech

POST

/v2/workspaces/{workspace\_id}/base-models/sam-audio-separate/inferences

```bash
curl --request POST \
  --url https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "prompt": "a knight standing in a snowy forest",
  "guidance_file_reference_audio": [
    {
      "url": "<file_url>"
    }
  ]
}'
```

```python
import requests


url = "https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences"


payload = {
    "prompt": "a knight standing in a snowy forest",
    "guidance_file_reference_audio": [{ "url": "<file_url>" }]
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}


response = requests.post(url, json=payload, headers=headers)


print(response.json())
```

```js
const url = 'https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences';
const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: '{"prompt":"a knight standing in a snowy forest","guidance_file_reference_audio":[{"url":"<file_url>"}]}'
};


try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}
```

```go
package main


import (
  "fmt"
  "strings"
  "net/http"
  "io"
)


func main() {


  url := "https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences"


  payload := strings.NewReader("{\n  \"prompt\": \"a knight standing in a snowy forest\",\n  \"guidance_file_reference_audio\": [\n    {\n      \"url\": \"<file_url>\"\n    }\n  ]\n}")


  req, _ := http.NewRequest("POST", url, payload)


  req.Header.Add("Authorization", "Bearer <token>")
  req.Header.Add("Content-Type", "application/json")


  res, _ := http.DefaultClient.Do(req)


  defer res.Body.Close()
  body, _ := io.ReadAll(res.Body)


  fmt.Println(res)
  fmt.Println(string(body))


}
```

```php
<?php


$curl = curl_init();


curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'prompt' => 'a knight standing in a snowy forest',
    'guidance_file_reference_audio' => [
        [
                'url' => '<file_url>'
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);


$response = curl_exec($curl);
$err = curl_error($curl);


curl_close($curl);


if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}
```

```java
OkHttpClient client = new OkHttpClient();


MediaType mediaType = MediaType.parse("application/json");
RequestBody body = RequestBody.create(mediaType, "{\n  \"prompt\": \"a knight standing in a snowy forest\",\n  \"guidance_file_reference_audio\": [\n    {\n      \"url\": \"<file_url>\"\n    }\n  ]\n}");
Request request = new Request.Builder()
  .url("https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences")
  .post(body)
  .addHeader("Authorization", "Bearer <token>")
  .addHeader("Content-Type", "application/json")
  .build();


Response response = client.newCall(request).execute();
```

```ruby
require 'uri'
require 'net/http'


url = URI("https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences")


http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true


request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"prompt\": \"a knight standing in a snowy forest\",\n  \"guidance_file_reference_audio\": [\n    {\n      \"url\": \"<file_url>\"\n    }\n  ]\n}"


response = http.request(request)
puts response.read_body
```

```csharp
var client = new RestClient("https://api.app.layer.ai/api/v2/workspaces/:workspace_id/base-models/sam-audio-separate/inferences");
var request = new RestRequest("", Method.Post);
request.AddHeader("Authorization", "Bearer <token>");
request.AddHeader("Content-Type", "application/json");
request.AddParameter("application/json", "{\n  \"prompt\": \"a knight standing in a snowy forest\",\n  \"guidance_file_reference_audio\": [\n    {\n      \"url\": \"<file_url>\"\n    }\n  ]\n}", ParameterType.RequestBody);
var response = client.Execute(request);
```

* Production

## Authorizations

* **[bearerAuth](/docs/v2/rest-api#bearerauth)**

## Parameters

### Path Parameters

**workspace\_id**

required

_Workspace Id_

Id of the workspace that owns the resource.

string format: uuid

Id of the workspace that owns the resource.

### Query Parameters

**session\_name**

Any of:

**string**

string

**null**

null

Organize this run under a named session. A new session is created if none matches.

## Request Bodyrequired

_SAM Audio Separate_object

**prompt**

_Prompt_

string

""

**guidance\_file\_reference\_audio**

_Reference audio_

Guides style and voice without being modified

Array<object>

_InferenceFormFileRef_

A single guidance/reference file supplied to a `reference_image_list` field.

The slot’s guidance `type` is fixed by the field it’s attached to; the caller supplies the file URL and, where the model supports per-file weighting, a weight.

object

**url**

required

_Url_

string

**weight**

_Weight_

number

nullable

**prompt\_language**

_Prompt Language_

_Advanced field._

string

nullable

##### Example

```json
{
  "prompt": "a knight standing in a snowy forest",
  "guidance_file_reference_audio": [
    {
      "url": "<file_url>"
    }
  ]
}
```

## Responses

### 202

Successful Response

_ExecuteForgeOutput_object

**inference\_id**

required

_Inference Id_

Unique identifier for this forge run.

string format: uuid

**status**

required

_InferenceStatus_

Current status: IN\_PROGRESS.

string

Allowed values: in\_progress complete failed cancelled deleted

**estimated\_price\_creative\_units**

Any of:

**number**

number

**null**

null

**poll\_interval\_seconds**

required

_Poll Interval Seconds_

Suggested polling interval in seconds.

integer

**created\_at**

required

_Created At_

Timestamp of when the run was created.

string format: date-time

**session\_id**

Any of:

**string**

string format: uuid

**null**

null

**normalized\_parameters**

Any of:

**NormalizedInferenceParameters**

_NormalizedInferenceParameters_

Inference parameters after model-specific normalization.

These reflect the actual values used for generation, including model defaults applied for any parameters not explicitly set.

object

**width**

Any of:

**null**

integer

**integer**

null

**height**

Any of:

**null**

integer

**integer**

null

**batch\_size**

Any of:

**null**

integer

**integer**

null

**num\_inference\_steps**

Any of:

**null**

integer

**integer**

null

**guidance\_scale**

Any of:

**null**

number

**number**

null

**duration\_seconds**

Any of:

**null**

number

**number**

null

**fps**

Any of:

**null**

integer

**integer**

null

**null**

null

##### Example

```json
{
  "status": "in_progress"
}
```

### 401

Unauthenticated — missing or invalid Bearer token.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```

### 403

Forbidden — insufficient permissions.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```

### 404

Resource not found.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```

### 422

Invalid input parameters.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```

### 429

Rate limited — too many concurrent requests.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```

### 500

Internal server error.

object

**type**

required

string

**title**

required

string

**status**

required

integer

**detail**

required

string

##### Example

```json
{
  "type": "https://api.layer.ai/errors/ERROR_CODE",
  "title": "Error Title",
  "status": 400,
  "detail": "Human-readable description."
}
```