Aria
AI Assistant
“Hi, I'm Aria. How can I help you today?”
AI assistantsProductivity
API v1 — live now
Real-time speech-to-text and voice workflows — live in minutes, not days.
Real-time transcription, voice cloning, and lifelike text-to-speech — over a small REST + WebSocket API. One key unlocks both products, with sub-second latency and drop-in compatibility for Whisper and ElevenLabs, so your team integrates once and ships faster everywhere.
Prefer Python? The official SDK wraps the ASR API with a fully-typed, async-friendly client.
Quickstart
Two env vars, two endpoints. You're live on both products with the same key.
# 1. Set your key + base URL once
export USF_API_KEY="<the key we sent you>"
export USF_BASE_URL="https://api-prod-usf.us.inc"
# 2. Transcribe a file
curl -sS "$USF_BASE_URL/v1/audio/transcriptions" \
-H "Authorization: Bearer $USF_API_KEY" \
-F file=@audio.wav \
-F model=usf-mini-asr
# 3. Or synthesize speech
curl -sS "$USF_BASE_URL/v1/text-to-speech/aria-ai-assistant?output_format=mp3_44100_128" \
-H "Authorization: Bearer $USF_API_KEY" \
-H "Content-Type: application/json" \
-d '{"text":"Hello from UltraSafe.","model_id":"usf-tts"}' \
-o speech.mp3A voice tuned to your product.
15 production-ready voices, each shaped for a specific use case — from healthcare to fintech, retail to news. Click any to hear it now, or clone your own from a 30-second sample.
Charles
Banking Advisor
“Good afternoon. I'd be happy to review your portfolio.”
FintechWealth management
Emma
Bedside Nurse
“Take a deep breath. I'm right here whenever you need me.”
HealthcarePatient support
Robert
News Anchor
“Tonight at six: a closer look at today's top stories.”
NewsEditorial
Priya
Meditation Guide
“Breathe in slowly. Hold. And gently exhale.”
WellnessMindfulness
Marco
Luxury Concierge
“Welcome. May I arrange anything to make your stay perfect?”
HospitalityPremium
Kai
Fitness Coach
“Let's go! Five more reps. You've got this!”
FitnessSports
Hannah
Online Tutor
“Great work! Let's move on to the next exercise.”
E-learningTutorials
Showing 8 of 15 built-in voices. Plus voice cloning from a 30-second sample.
Try the full catalogueTwo products. One integration. No surprises.
A handful of REST endpoints and a single WebSocket — documented in detail, with copy-pastable examples in curl, Python, and JavaScript. Same auth, same base URL, same SLA across both speech-to-text and text-to-speech.
Speech-to-text
Stream microphone audio with sub-second latency or transcribe whole files. Same model, same accuracy across both paths — plus VAD, diarization, and translation, each opt-in per request.
Text-to-speech
Lifelike voices for assistants, IVR, and prompts. 15+ built-in voices, voice cloning from a 30-second sample, and streaming output in MP3, PCM, or µ-law.
One key, both products
A single API key unlocks transcription, translation, and synthesis. No separate billing, no separate dashboards, no second integration to maintain.
Built for production
Dedicated GPUs in Stockholm with autoscaling. Typical 100–200 ms end-to-end for short utterances on either product, with steady throughput under load.
Per-tester keys, audited
Every key is scoped to one tester and revocable in seconds. Every request is logged with user, IP, route, and status — so you always know who hit what.
Drop-in compatible
Whisper-style /v1/audio/* and ElevenLabs-style /v1/text-to-speech/* endpoints. Point existing client code at this base URL — most stacks just work.
Voice and speech, live in 60 seconds.
Get an API key by email, copy a curl from the docs, and ship transcription or text-to-speech into your product today. Real-time streaming, voice cloning, and translation are all one parameter away.