Blog

Engineering insights from the Reson8 team

Rumble-2 written in white on a deep berry background
Launch

Meet Rumble-2

Our most accurate speech-to-text model, now live in 50+ languages. Lowest streaming WER and p95 latency on English, and ahead of medical-specific models.

Reson8 x Bonnie
Customer Story

How Bonnie captures phone numbers and email addresses during calls

A voice agent for more than 1,000 restaurants that no longer cuts callers off mid-sentence and gets contact details right the first time, on the call.

Rumble-2-flash written in white on a green background
Launch

Meet Rumble-2-flash

Our smallest, fastest and most affordable model yet. It understands 40+ languages, starting at €0.01 per hour of audio, and it's live today.

A reference transcript aligned word by word with a model transcript, showing one substituted word and one deleted word
Guide

How to benchmark speech-to-text on your own audio

What to measure, how to make your test set sound like production, why reference transcripts are less reliable than you think, and how much audio you need.

Reson8 x Autoscriber
Customer Story

How Autoscriber gets medical terminology and dosages right

Medication names, dosages and readings that come back right in four languages, with 62% lower word error rate on mixed-language audio and transcripts five times faster.

Sifted logo for the Sifted AI 100 2026
News

Reson8 featured in Sifted AI 100 2026

#82 on Sifted's ranking of Europe's most promising AI startups in 2026. Founded this year, and at 10 people the leanest company on the list.

Reson8 x Jack & Jill
Customer Story

How Jack & Jill improved Jack's conversations by halving turns per call

End-of-turn detection that lets candidates finish their thought, with 50% longer utterances and full rollout 15 days after the first meeting.

Reson8 x SafeGPT
Customer Story

How SafeGPT keeps Dutch care and government audio in the Netherlands

No cloud provider in the processing chain, no audio retained, and inference on hardware Reson8 owns and runs in a Dutch data center.

Reson8 x VEED
Customer Story

How VEED keeps subtitles in sync to the word

Word-level timestamps precise enough to display on and to edit on, tuned to the way VEED benchmarks every provider it runs.

EU circle of stars around a sound wave, representing speech data kept in the EU
Engineering

Keeping audio in the EU: why we run our own GPUs

An EU region only answers where your audio sits. Owning the hardware in the Netherlands answers who owns it, whose law applies and what is kept.

Reson8 x Obsidian
Customer Story

How Obsidian gets the numbers right

Accurate transcription of adviser meetings, with Reson8 self-hosted in Obsidian's own AWS environment.

Focus brackets locking onto the identifier INV-2048 between two sound waves
Product

Introducing Patterns

Bias the model toward the structured identifiers your transcripts depend on — order codes, invoice numbers, licence plates — and cut identifier error by up to 81%.

One speaker's sound wave hands over to the next at an end-of-turn marker
Product

Introducing the Turns endpoint

Eager, accurate turn detection for voice agents, with native barge-in — so your agent knows when to listen, when to speak, and when to get out of the way.

Resonant-1 written in white on a blue background
Benchmarks

Introducing Resonant-1 and Resonant-1-flash

#1 Open ASR Leaderboard
6.51% Average WER
1438x Realtime (Flash)

Two frontier speech models trained with reinforcement learning, served via Violin — our inference engine built for speech.