Language selection - AssemblyAI

Pass language_codes in the config part to steer transcription toward a specific language. The field always takes a list, so use a single-element list such as ["es"] for monolingual audio, and several codes for multilingual or code-switching audio.

Accepts ISO 639-1 codes in a list: ["es"], or ["en", "es"]
Defaults to English (en) when omitted
Applies to the transcript only. The rewrite in llm_response follows the language of the transcript unless your llm_instruction says otherwise.

Python SDK
Python
JavaScript
cURL

import assemblyai as aai

aai.settings.api_key = "<YOUR_API_KEY>"

config = aai.DictationConfig(language_codes=["es"])

result = aai.DictationTranscriber().transcribe_live("clip.wav", config)
print(result.text)

import json
import requests

config = {"language_codes": ["es"]}

with open("clip.wav", "rb") as f:
    audio = f.read()

response = requests.post(
    "https://dictation.assemblyai.com/v1/transcribe/live",
    headers={"Authorization": "<YOUR_API_KEY>"},
    files={
        "config": (None, json.dumps(config), "application/json"),
        "audio": ("clip.wav", audio, "audio/wav"),
    },
    timeout=90,
)
response.raise_for_status()
print(response.json()["text"])

import { readFileSync } from "fs";

const config = { language_codes: ["es"] };

const audio = readFileSync("clip.wav");
const form = new FormData();
form.append(
  "config",
  new Blob([JSON.stringify(config)], { type: "application/json" })
);
form.append("audio", new Blob([audio], { type: "audio/wav" }), "clip.wav");

const response = await fetch(
  "https://dictation.assemblyai.com/v1/transcribe/live",
  { method: "POST", headers: { Authorization: "<YOUR_API_KEY>" }, body: form }
);
console.log((await response.json()).text);

curl -X POST https://dictation.assemblyai.com/v1/transcribe/live \
  -H 'Authorization: <YOUR_API_KEY>' \
  -F 'config={"language_codes": ["es"]};type=application/json' \
  -F 'audio=@clip.wav;type=audio/wav'

Multilingual audio

Pass several codes when a single clip moves between languages, which is common in dictation from bilingual speakers:

{ "language_codes": ["en", "es"] }

List only the languages you actually expect. Extra codes give the model more ways to be wrong on ambiguous audio.

Supported languages

Code	Language	Code	Language
`en`	English	`da`	Danish
`es`	Spanish	`fi`	Finnish
`de`	German	`hi`	Hindi
`fr`	French	`vi`	Vietnamese
`it`	Italian	`ar`	Arabic
`pt`	Portuguese	`he`	Hebrew
`tr`	Turkish	`ja`	Japanese
`nl`	Dutch	`ur`	Urdu
`sv`	Swedish	`zh`	Mandarin
`no`	Norwegian

Prompting and keyterms steers the transcript within the chosen language.
Transcript rewriting reshapes the transcript after it is written.