Voice AI · REST API

Speech, decoded.

Fast, accurate speech‑to‑text.

Production‑grade speech AI in a single API call. Get a key and ship — no setup, no card.

bash — daglo-stt

# transcribe an audio file $ curl -X POST https://apis.daglo.ai/stt/v1/async/transcripts \ -H "Authorization: Bearer $DAGLO_KEY" \ -F "file=@meeting.mp3"

{ "status": "transcribed", "confidence": 0.95, "text": "Let's go over today's meeting agenda." }

95%+

transcription accuracy

4 hrs

max audio per file

Stream + Batch

live & file‑upload ASR

// Endpoints

Speech‑to‑Text · v1

POST

/stt/v1/async ★

Transcribe long audio — up to 4 hours / 2 GB per file.

POST

/stt/v1/sync

Instant transcription for clips under 30 seconds.

STREAM

/stt/v1/stream

Real‑time speech‑to‑text from a live mic stream (gRPC).

CONFIG

diarization

Speaker diarization — label who said what.

CONFIG

keywordBoost

Boost recognition of your domain keywords.

LANG

ko · en · ja

Korean, English, Japanese, and KO+EN mixed.

Fast

Low‑latency streaming and batch. Stream partials as you speak, or process files at scale.

Accurate

Models tuned for real conversations — noise, accents, and domain terms included.

Simple

One REST call, clear JSON, predictable pricing. SDKs and a live Console to ship faster.

Talk to us

Questions about the API, pricing, or on‑premise? Send us your inquiry and we will reply within 1 business day.

Partnershippartnership@actionpower.kr