ollalink API

What is ollalink?

Service status (updated 2026-11-03)

Available: file transcription and speech detection, batch speech-to-text for Hindi, Hinglish and other languages, realtime speech-to-text and live captions, text translation, text-to-speech, text tools (spoken numbers to digits, embeddings, classification), chat and coding, vision and document OCR, image generation and editing, audio generation and music, and speaker diarization. When the live speech lanes are full, a new session is refused with busy and close code 1013 — retry shortly. New: voice design in the 14 languages. Paused: voice cloning and sound-stream. Host names, routes and keys are unchanged.

ollalink is a partner API for speech. Text-to-speech, chat and coding are OpenAI-compatible, so an OpenAI SDK works by changing the base URL and adding the key header; the other capabilities have their own documented routes.

Speech recognition and text translation cover 14 languages: English, Hindi, Spanish, French, Chinese (Mandarin), German, Arabic, Brazilian Portuguese, Russian, Japanese, Korean, Italian, Dutch and Vietnamese; text translation also covers Tamil. Text-to-speech covers 6 languages: English, Spanish, French, German, Brazilian Portuguese and Hindi. See Models & languages for the per-lane scope and Speech routing before planning a large event.

Speech in

Transcribe audio in 14 languages — batch for Hindi and Hinglish, fast file transcription with subtitles, and a realtime streaming lane. See Speech-to-text.

Live captions and live translation

Stream a speaker’s audio over a WebSocket and get live captions in the speaker’s language; send each final caption to the translation API for live translation into any of the 14 languages or Tamil. See Live captions.

Speech out

Text-to-speech with 34 voices in English, Spanish, French, German, Brazilian Portuguese and Hindi, or design a voice from a description in any of the 14 languages. See Text-to-speech.

Translation and text tools

Translate text between the 14 languages and Tamil with numbers, IP addresses and commands kept intact, turn spoken numbers into digits (“port four forty three” → 443), create text embeddings and classify text. See Translation and text tools.

Chat, coding and more

Chat and coding models that are always on, plus vision, document OCR and speaker diarization. See Chat and coding and Models.

Images, sound and music

Generate and edit images from a text prompt, create sound effects and instrumental music clips, and compose songs from lyrics. See Models.

Services

Service	Base host	Auth	Covers
Translation, text tools + models	https://ai.ollalink.com	X-NH-GPU-Key: <key>	Text translation (POST /v1/translate), spoken numbers to digits (POST /v1/normalize), embeddings (POST /v1/embeddings), classification (POST /v1/classify), chat, coding and vision (/v1/chat/completions), document OCR (/v1/ocr), images (/v1/images/generations, /v1/images/edits), audio (/v1/audio/generations), music (/v1/music/generations)
Speech GPU	gpu-*.ollalink.com	X-NH-GPU-Key: <key>	Batch, file and realtime speech-to-text · speaker diarization · live captions · text-to-speech

One GPU API key from the dashboard opens both. Voice cloning and the realtime speech-to-speech service are not available; the live caption lane returns captions in the speaker’s language, and translation is a separate call (what is not offered).

Start with Getting started, then Authentication.