Build with the
languu API.

Authentication

Every request authenticates with a secret API key sent as a Bearer token. Generate keys in the dashboard under Settings → API keys. Keep secret keys server-side; never ship a lang_sk_ key to the browser.

curl https://api.languu.com/v1/translate \
-H "Authorization: Bearer lang_sk_live_xxx" \
-H "Content-Type: application/json" \
-d '{ "text": "Hello world", "target": "ko" }'

Errors & limits

The API uses conventional HTTP status codes. Failures return a JSON body with a stable code string you can branch on and a human-readable message for logs.

Default limits are 60 requests/minute for text endpoints and 10 concurrent jobs for media. Responses include X-RateLimit-Remaining and Retry-After headers. Need more? Contact support to lift limits per workspace.

Translation

Translate text between 30+ languages with tone and glossary controls. Send text (string or array) plus a target language; omit source to auto-detect.

const res = await fetch("https://api.languu.com/v1/translate", {
method: "POST",
headers: {
"Authorization": `Bearer ${process.env.LANGUU_KEY}`,
"Content-Type": "application/json"
},
body: JSON.stringify({
text: "The quick brown fox.",
target: "ko", // Korean
formality: "formal" // "formal" | "casual" | "default"
})
});

const { translations } = await res.json();
console.log(translations[0].text);

Pass a glossary_id to enforce brand terms, or send text as an array to translate up to 100 strings in one call while preserving order.

Transcription

Convert audio or video to time-stamped text. Submit a publicly reachable media_url (or upload via the signed-URL flow) and poll the returned job until status is completed.

curl https://api.languu.com/v1/transcribe \
-H "Authorization: Bearer $LANGUU_KEY" \
-H "Content-Type: application/json" \
-d '{
  "media_url": "https://cdn.example.com/interview.mp3",
  "language": "auto",
  "diarize": true,
  "timestamps": "word"
}'

# → { "id": "txn_8fa2", "status": "processing" }

Set diarize: true to label speakers and timestamps to "word" or "segment". Provide a webhook_url and skip polling — languu posts the finished transcript to your endpoint.

Text to speech

Generate natural speech from text. Choose a voice, optionally tune speed and format, and stream or download the audio.

const res = await fetch("https://api.languu.com/v1/speech", {
method: "POST",
headers: {
"Authorization": `Bearer ${process.env.LANGUU_KEY}`,
"Content-Type": "application/json"
},
body: JSON.stringify({
text: "Welcome to languu.",
voice: "aria", // list voices at /v1/voices
format: "mp3", // "mp3" | "wav" | "ogg"
speed: 1.0
})
});

const audio = await res.arrayBuffer();

For long passages, send stream: true to receive chunked audio as it renders. SSML is supported in the text field for pauses, emphasis, and pronunciation.

Closed captioning

Produce broadcast-ready captions from media in a single job. Captioning chains transcription and (optionally) translation, then returns subtitle files in the formats you request.

curl https://api.languu.com/v1/captions \
-H "Authorization: Bearer $LANGUU_KEY" \
-H "Content-Type: application/json" \
-d '{
  "media_url": "https://cdn.example.com/lecture.mp4",
  "formats": ["srt", "vtt"],
  "translate_to": ["es", "ja"],
  "max_chars_per_line": 42
}'

The completed job returns a download URL per format per language. Use max_chars_per_line and max_lines to match your platform's caption style guide.

API reference

All endpoints are rooted at https://api.languu.com/v1 and accept and return JSON unless noted.

Retrieve the status and result of any media job (transcribe or captions). Returns status of processing, completed, or failed, plus the output payload once finished.

Mock-mode runbook

Mock mode runs the entire languu frontend against deterministic local fixtures instead of the AWS backend. Use it to preview the UI, develop offline, run UI tests in CI, and demo flows without API keys, network access, or billing.

Enable mock mode

1. Install dependencies

   From the project root, install once.

   bashCopy

   cd frontend && npm install
   cd ../backend && npm install

2. Set the environment flag

   Add this to frontend/.env.local for in-browser mocks, or use the local mock API URL.

   env · .env.localCopy

   # In-browser mocks only:
   NEXT_PUBLIC_API_URL=mock
   
   # Or local mock REST API (run: cd backend && npm run local:mock-api):
   # NEXT_PUBLIC_API_URL=http://127.0.0.1:3333

3. Run the dev server

   Start as usual — no live API key required when using mock mode.

   bashCopy

   cd frontend
   npm run dev
   # ▲ ready on http://localhost:3000

How fixtures resolve

In-browser mock mode short-circuits API calls with stable sample responses. The local mock API serves the same shapes from backend/scripts/local-rest-api.ts. Media jobs return processing on creation, then completed on poll.

Switching back to live

Set NEXT_PUBLIC_API_URL to your deployed API Gateway URL from CDK output. Authenticate with Cognito via the app login flow — no code changes required.