Speech recognition that adapts to your language in seconds

Try for free
Book a call

Built for production

When accuracy, reliability and privacy are non-negotiable.

Accurate on domain‑specific language

Medication names, order numbers, addresses, or case references - all transcribed accurately. Adapts from a text upload or in real-time, no audio samples or fine tuning.

Native-grade transcription across European languages

English, Dutch, French, German, Italian, Polish, (Brazilian) Portuguese, Spanish, Swedish, Frisian.

EU native, zero audio retention

Built for regulated industries, with 100% GPU processing in EU, zero audio retention, and no customer data used for training. ISO 27001 certified, GDPR and EU AI Act compliant.

EU AI Act compliantGDPR compliantISO 27001 certified
View trust center

An endpoint for any audio

PRERECORDED

Batch transcription for the highest accuracy on complete audio files

REALTIME

Streaming transcription for live conversations and dictation

TURNS

Turn-aware transcription with barge-in detection for voice agents

Healthcare voice engine

Healthcare
automotive
Legal
finance
customer service
Supports 50,000+ medical terms with no latency impact
Reliable transcription in real-world driving conditions
Accurate recognition across complex legal vocabulary
High precision for numbers, tickers, and financial entities
Clear intent detection at scale

No tradeoffs between
accuracy and speed

Generic speech engines fail on domain‑specific language.
Reson8 gets the words right that others get wrong.

Reson8 adapts to your business context in real-time or with a text upload. No audio files or finetuning required.

Accurate transcription of order codes, reference numbers, licence plates, invoice numbers, flight numbers, and more.

Smaller, faster models give you infrastructure which works in the real world.

New

Rumble-2,
for production

Industry-leading accuracy for enterprises, across 50+ languages.

Available on Prerecorded, Realtime, and Turns.

Zero data retention, runs on our own GPUs in the Netherlands.

Includes automatic language detection and custom models. Diarization available.

explore rumble-2

New

Rumble-2-Flash,
for builders

Transcription from €0.01 per hour of audio, or free to run yourself.

Our fastest model: an hour of audio in seconds, not minutes.

Covers 40+ languages.

Hosted inference via our API, or open weights under Apache 2.0.

Read the announcement
View on Hugging Face

PRICING

From pilots to production deployments

Previous plans
Next plans

Build

€0
/ mo
Free credits equivalent to
~12 h Prerecorded
or ~8 h Realtime / Turns
  • 1 concurrent connection
  • Basic support
  • No card required, upgrade any time
Get started

Launch

€20
/ mo
Credits equivalent to
~125 h Prerecorded
or ~83 h Realtime / Turns
  • 20 concurrent connections
  • Standard support
  • Extra usage at the same rates, no cap
Get started
Recommended

Scale

€100
/ mo
Credits equivalent to
~714 h Prerecorded
or ~476 h Realtime / Turns
  • 100 concurrent connections
  • Priority support
  • Extra usage at the same rates, no cap
  • 12.5% lower rates versus Launch plan
Get started

Enterprise

Custom
Volume pricing
Annual contracts
from €10k / year
  • Custom concurrency
  • Dedicated support and SLA
  • Committed volume, custom rates
  • On-prem / VPC deployment
Talk to us
View pricing breakdown

Frequently asked questions

Reson8 is built for European languages and gets the words right that matter most. Generic speech models mishear medication names, order numbers, addresses and case references. Reson8’s speech-to-text adapts to your domain-specific terminology in real time or from a single text upload, no fine-tuning or audio samples needed.

We have full support for 50+ languages. See more here.

Reson8 routinely leads speech-to-text benchmarks from third-parties such as Hugging Face, Artificial Analysis, Coval, and Voice Arena.

Yes, a single stream can switch dynamically between languages mid-conversation. Language detection is automatic and out-of-the-box, no manual language flagging necessary. Find more information on language detection here.

Yes, diarization is supported on our Prerecorded and Realtime endpoints, ensuring transcripts accurately distinguish between speakers no matter the complexity of the audio. Find more information on diarization here.

Each endpoint is applicable for a different use case:

  • Prerecorded: for complete audio files, uploaded over REST. You get the full transcript back in a single response.
  • Realtime: for live audio streamed over WebSocket, when you want to show live transcripts while speaking.
  • Turns: also a WebSocket stream, but it detects when a speaker starts and stops talking, returning the transcript of each turn separately. Use this endpoint for cases like voice agents, when you need low latency end-of-turn detection and native barge-in.

Check out our guide on choosing the right endpoint here.

No, we don't and never will. Audio is processed on our own GPU infrastructure, which is fully built and operated by Reson8. Find more information on our trust center here.

Reson8 retains zero data. All audio is streamed pass-through and never stored; this also applies to files and transcripts.

Yes, Reson8 is fully ISO 27001 certified and GDPR compliant. We own and operate our full GPU infrastructure ourselves in the Netherlands, Reson8 is not reliant on any third party subprocessors of audio, making Reson8 uniquely suitable to regulated industries like healthcare, legal, finance, and more. Find our DPA here.

Sign up today! Get an API key through console.reson8.dev, talk to our team or check our documentation to get started!

Yes. Rumble-2-Flash is available as open weights under Apache 2.0, so you can download it and run it on your own servers or device. For Rumble-2, we offer on-prem and VPC deployment on the Enterprise plan.

voice ai
that speaks
your language

Try for Free
Book a call