> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nyra-labs.com/api-reference/audio/transcribe/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nyra-labs.com/_mcp/server.

# Create transcription

POST https://api.nyra-labs.com/v1/audio/transcriptions
Content-Type: multipart/form-data

Transcribes a recording and answers with two transcripts: `verbatim`, what was said, including fillers, repetitions and cut-offs; and `intended`, what was meant, as fluent text. Both carry word-level timestamps, and `alignment` relates the two word by word. Send `language` when you know it. The [quickstart](/quickstart) walks through the request, the accepted formats, the [supported languages](/quickstart#supported-languages) and every field of the response.

Reference: https://docs.nyra-labs.com/api-reference/audio/transcribe

## Authentication

- `Authorization` header (bearer token, required) — Your API key, e.g. `Authorization: Bearer nl_live_...`.

## Servers

- `https://api.nyra-labs.com` (Production, default)
- `http://localhost:3001` (Local)

## Request

### Body (multipart/form-data)

This endpoint expects a multipart form containing a file.

- `file` (file, required) — The recording to transcribe, up to 25 MB. Accepted formats, detected from the bytes rather than the filename: wav, mp3, m4a (AAC in an MP4 container), flac, ogg (Vorbis or Opus) and webm. Video files are refused; send the audio track.
- `language` (enum, optional) — The language of the audio as an ISO-639-1 code, e.g. `en` or `de`. Optional but recommended: when it is omitted the language is detected from the first 30 seconds, which costs a little time and can guess wrong on short or noisy recordings. See the [supported languages](/quickstart#supported-languages): German, English, Spanish, French, Portuguese, Russian, Swedish, Dutch, Polish, Italian, Danish and Ukrainian are officially supported; the other codes are available in beta. A code outside the list answers `unsupported_language`.
- `model` (enum, required) — The model to transcribe with. Required; `crisperwhisper-v2` is the only model, and any other value answers `model_not_found`.

## Response

### 200

The transcription.

- `alignment` (list of AlignmentEntry, required, nullable) — The verbatim transcript aligned against the intended one, in reading order. `null` when the two could not be aligned; the transcripts are still valid.
- `duration` (double, required) — Duration of the decoded audio, in seconds.
- `intended` (Transcript, required) — What the speaker meant: fluent, without fillers, repetitions or cut-offs.
- `language` (string, required) — ISO-639-1 code of the audio's language, as sent or as detected. See the [supported languages](/quickstart#supported-languages): twelve are officially supported, the rest are available in beta.
- `usage` (Usage, required)
- `verbatim` (TranscriptionVerbatim, required) — What the speaker said, including fillers, repetitions and cut-offs.

## Errors

### 400 Bad Request Error

The body is not multipart (`invalid_content_type`); `file` or `model` is missing (`missing_required_parameter`); the model does not exist (`model_not_found`); the language is not supported (`unsupported_language`); a removed parameter was sent (`unsupported_parameter`); the file is empty or not audio (`invalid_audio`), not an accepted format or a video (`unsupported_file_type`), or longer than an hour (`audio_too_long`).

- `error` (ErrorResponseError, required)

### 401 Unauthorized Error

The API key is missing, malformed or invalid.

- `error` (ErrorResponseError, required)

### 403 Forbidden Error

The API key is missing the `transcriptions:write` scope.

- `error` (ErrorResponseError, required)

### 413 Content Too Large Error

The upload exceeds the 25 MB limit.

- `error` (ErrorResponseError, required)

### 429 Too Many Requests Error

Rate limit exceeded, or the organization's wallet balance cannot cover the request (`insufficient_quota`).

- `error` (ErrorResponseError, required)

### 503 Service Unavailable Error

The transcription backend is at capacity. Retry after the `retry-after` header says.

- `error` (ErrorResponseError, required)

## Types

### AlignmentEntry

- `intended` (string, required, nullable) — The intended-transcript token, or `null` when nothing was meant here.
- `type` (enum, required) — `match`: said and meant. `repetition`: a repeated word. `filler`: um, uh and their kin. `cutoff`: a word broken off. `sound`: a bracketed non-speech event. `disfluent`: verbatim speech with no intended counterpart. `correction`: the verbatim token differs from what was meant.
  - Allowed values: `match`, `repetition`, `disfluent`, `cutoff`, `filler`, `sound`, `correction`
- `verbatim` (string, required, nullable) — The verbatim-transcript token, or `null` when nothing was said here.

### Transcript

What the speaker meant: fluent, without fillers, repetitions or cut-offs.

- `text` (string, required)
- `words` (list of TranscriptWord, required) — Every word, in order, with timings.

### Usage

- `seconds` (double, required) — Billed audio seconds: the audio's duration rounded up to a whole second.
- `type` (enum, required)
  - Allowed values: `duration`

### TranscriptionVerbatim

What the speaker said, including fillers, repetitions and cut-offs.

- `text` (string, required)
- `words` (list of TranscriptWord, required) — Every word, in order, with timings.

### ErrorResponseError

- `code` (string, required, nullable)
- `message` (string, required)
- `param` (string, required, nullable)
- `type` (enum, required)
  - Allowed values: `invalid_request_error`, `authentication_error`, `permission_error`, `rate_limit_error`, `insufficient_quota`, `server_error`

### TranscriptWord

- `alignment_index` (integer, required, nullable) — Index into `alignment` of the entry this word belongs to, or `null` when the link could not be established. Never guessed.
- `end` (double, required) — End of the word, in seconds.
- `start` (double, required) — Start of the word, in seconds.
- `word` (string, required) — The word, keeping its leading space, so that joining every `word` of a transcript reproduces its `text`.

## Examples

**Request**

```json
{
  "file": "<file: string>",
  "model": "crisperwhisper-v2"
}
```

**Response**

```json
{
  "alignment": [
    {
      "intended": null,
      "type": "repetition",
      "verbatim": "I"
    },
    {
      "intended": "I",
      "type": "match",
      "verbatim": "I"
    },
    {
      "intended": "want",
      "type": "match",
      "verbatim": "want"
    },
    {
      "intended": null,
      "type": "filler",
      "verbatim": "um"
    },
    {
      "intended": "to",
      "type": "match",
      "verbatim": "to"
    },
    {
      "intended": "go",
      "type": "match",
      "verbatim": "go"
    },
    {
      "intended": "home",
      "type": "match",
      "verbatim": "home"
    }
  ],
  "duration": 2.84,
  "intended": {
    "text": "I want to go home",
    "words": [
      {
        "alignment_index": 1,
        "end": 0.48,
        "start": 0.34,
        "word": " I"
      },
      {
        "alignment_index": 2,
        "end": 0.81,
        "start": 0.52,
        "word": " want"
      },
      {
        "alignment_index": 4,
        "end": 1.69,
        "start": 1.58,
        "word": " to"
      },
      {
        "alignment_index": 5,
        "end": 1.94,
        "start": 1.72,
        "word": " go"
      },
      {
        "alignment_index": 6,
        "end": 2.41,
        "start": 1.98,
        "word": " home"
      }
    ]
  },
  "language": "en",
  "usage": {
    "seconds": 3,
    "type": "duration"
  },
  "verbatim": {
    "text": "I I want um to go home",
    "words": [
      {
        "alignment_index": 0,
        "end": 0.26,
        "start": 0.12,
        "word": " I"
      },
      {
        "alignment_index": 1,
        "end": 0.48,
        "start": 0.34,
        "word": " I"
      },
      {
        "alignment_index": 2,
        "end": 0.81,
        "start": 0.52,
        "word": " want"
      },
      {
        "alignment_index": 3,
        "end": 1.31,
        "start": 0.95,
        "word": " um"
      },
      {
        "alignment_index": 4,
        "end": 1.69,
        "start": 1.58,
        "word": " to"
      },
      {
        "alignment_index": 5,
        "end": 1.94,
        "start": 1.72,
        "word": " go"
      },
      {
        "alignment_index": 6,
        "end": 2.41,
        "start": 1.98,
        "word": " home"
      }
    ]
  }
}
```

**SDK Code**

```python
import requests

url = "https://api.nyra-labs.com/v1/audio/transcriptions"

files = { "file": "open('string', 'rb')" }
payload = {
    "language": ,
    "model": "crisperwhisper-v2"
}
headers = {"Authorization": "Bearer <token>"}

response = requests.post(url, data=payload, files=files, headers=headers)

print(response.json())
```

```javascript
const url = 'https://api.nyra-labs.com/v1/audio/transcriptions';
const form = new FormData();
form.append('file', 'string');
form.append('language', '');
form.append('model', 'crisperwhisper-v2');

const options = {method: 'POST', headers: {Authorization: 'Bearer <token>'}};

options.body = form;

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.nyra-labs.com/v1/audio/transcriptions"

	payload := strings.NewReader("-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"file\"; filename=\"string\"\r\nContent-Type: application/octet-stream\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"language\"\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"model\"\r\n\r\ncrisperwhisper-v2\r\n-----011000010111000001101001--\r\n")

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

	req.Header.Add("Authorization", "Bearer <token>")

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

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

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

}
```

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

url = URI("https://api.nyra-labs.com/v1/audio/transcriptions")

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

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request.body = "-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"file\"; filename=\"string\"\r\nContent-Type: application/octet-stream\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"language\"\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"model\"\r\n\r\ncrisperwhisper-v2\r\n-----011000010111000001101001--\r\n"

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

```java
import com.mashape.unirest.http.HttpResponse;
import com.mashape.unirest.http.Unirest;

HttpResponse<String> response = Unirest.post("https://api.nyra-labs.com/v1/audio/transcriptions")
  .header("Authorization", "Bearer <token>")
  .body("-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"file\"; filename=\"string\"\r\nContent-Type: application/octet-stream\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"language\"\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"model\"\r\n\r\ncrisperwhisper-v2\r\n-----011000010111000001101001--\r\n")
  .asString();
```

```php
<?php
require_once('vendor/autoload.php');

$client = new \GuzzleHttp\Client();

$response = $client->request('POST', 'https://api.nyra-labs.com/v1/audio/transcriptions', [
  'multipart' => [
    [
        'name' => 'file',
        'filename' => 'string',
        'contents' => null
    ],
    [
        'name' => 'model',
        'contents' => 'crisperwhisper-v2'
    ]
  ]
  'headers' => [
    'Authorization' => 'Bearer <token>',
  ],
]);

echo $response->getBody();
```

```csharp
using RestSharp;

var client = new RestClient("https://api.nyra-labs.com/v1/audio/transcriptions");
var request = new RestRequest(Method.POST);
request.AddHeader("Authorization", "Bearer <token>");
request.AddParameter("undefined", "-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"file\"; filename=\"string\"\r\nContent-Type: application/octet-stream\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"language\"\r\n\r\n\r\n-----011000010111000001101001\r\nContent-Disposition: form-data; name=\"model\"\r\n\r\ncrisperwhisper-v2\r\n-----011000010111000001101001--\r\n", ParameterType.RequestBody);
IRestResponse response = client.Execute(request);
```

```swift
import Foundation

let headers = ["Authorization": "Bearer <token>"]
let parameters = [
  [
    "name": "file",
    "fileName": "string"
  ],
  [
    "name": "language",
    "value": 
  ],
  [
    "name": "model",
    "value": "crisperwhisper-v2"
  ]
]

let boundary = "---011000010111000001101001"

var body = ""
var error: NSError? = nil
for param in parameters {
  let paramName = param["name"]!
  body += "--\(boundary)\r\n"
  body += "Content-Disposition:form-data; name=\"\(paramName)\""
  if let filename = param["fileName"] {
    let contentType = param["content-type"]!
    let fileContent = String(contentsOfFile: filename, encoding: String.Encoding.utf8)
    if (error != nil) {
      print(error as Any)
    }
    body += "; filename=\"\(filename)\"\r\n"
    body += "Content-Type: \(contentType)\r\n\r\n"
    body += fileContent
  } else if let paramValue = param["value"] {
    body += "\r\n\r\n\(paramValue)"
  }
}

let request = NSMutableURLRequest(url: NSURL(string: "https://api.nyra-labs.com/v1/audio/transcriptions")! as URL,
                                        cachePolicy: .useProtocolCachePolicy,
                                    timeoutInterval: 10.0)
request.httpMethod = "POST"
request.allHTTPHeaderFields = headers
request.httpBody = postData as Data

let session = URLSession.shared
let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in
  if (error != nil) {
    print(error as Any)
  } else {
    let httpResponse = response as? HTTPURLResponse
    print(httpResponse)
  }
})

dataTask.resume()
```