Managed transcription.
With speakers attributed.
Upload audio or stream a microphone — get back a timestamped transcript with each speaker labelled. Built on the same models as our open-source SDK, hosted for when you don’t want to ship the weights yourself.
Everything you need to ship voice features.
Speech-to-text, speaker separation, speaker identification, and OpenAI-compatible APIs — bundled into one price.
Upload an audio file for batch transcription, or stream a WebSocket for live captioning. Same authentication, same billing, same speaker handling.
Every utterance comes back attributed to a speaker. Register speaker profiles once and get stable named identities across meetings and call recordings.
Parakeet TDT for 25 European languages plus Meta Omnilingual for the long tail (1,672 languages including Hindi, Arabic, Indonesian, Vietnamese). Auto-detect or explicit language hint.
1¢ per audio-second on the published tier ($0.60/min for everything — speech-to-text, diarization, and speaker ID bundled). Volume discounts for paid tiers; enterprise SLAs available. No minimums.
Existing code written against the OpenAI Whisper API works against Soniqo by changing one configuration line (the base URL). No client rewrite, no SDK migration.
Sign in with Google or GitHub for the console, or generate long-lived API keys for server-to-server traffic. Argon2-hashed; revocable any time.
Durable queue with explicit acknowledgement and idempotency keys. A worker preemption never loses or double-bills a job — the ledger is the source of truth.
Transcripts and speaker profiles are tied to your account and exportable via the API. Self-serve account deletion. See the Privacy Policy for details.
One curl call to transcribe an audio file.
After signing in you'll have a starter balance and one API key. The endpoint is HTTP multipart upload; here's the smallest possible call.
# Batch transcription with diarization
curl -X POST https://api.soniqo.audio/v1/transcribe \
-H "Authorization: Bearer $SONIQO_API_KEY" \
-F "audio=@meeting.wav"
# Returns a job id; poll /v1/transcribe/<id> for the transcript.Already using the OpenAI SDK? Change the base URL and your existing code works:
# Drop-in OpenAI-Whisper-compatible endpoint
from openai import OpenAI
client = OpenAI(
base_url="https://api.soniqo.audio/v1",
api_key="<your-soniqo-api-key>",
)
with open("meeting.wav", "rb") as f:
transcript = client.audio.transcriptions.create(
model="whisper-1",
file=f,
)
print(transcript.text)One number covers everything.
Same models, same API — for free.
The cloud is for convenience. If you want full control or zero per-minute cost, run the open-source library on your own infrastructure. Apache 2.0, no per-minute pricing, nothing leaves your machines.
The same speech recognition, speaker separation, and voice synthesis models, packaged for Apple platforms. Swift API, Homebrew install, Apache 2.0.
A cross-platform C++ engine for Linux, Windows, macOS, and Android — speech-to-text, diarization, speaker ID, and realtime streaming (Nemotron Speech Streaming) on CPU. Apache 2.0.