# XMOS

> XMOS is a fabless semiconductor company based in Bristol, UK, that designs the XCORE family of multicore, deterministic processors for embedded voice, audio, AI and control at the edge. Founded in 2005 by a team rooted in the University of Bristol and Inmos transputer heritage, XMOS built its name on far-field voice processing (VocalFusion, the first far-field solution qualified for Amazon Alexa) and now positions XCORE as a 'Generative System-on-Chip' (GenSoC) - silicon whose behaviour can be described in natural language while guaranteeing real-time, deterministic performance.

- **Founded:** 2005
- **Headquarters:** Bristol, United Kingdom
- **Founders:** David May (Co-founder / CTO (former Inmos chief architect)), James Foster (Co-founder (former CEO of Oxford Semiconductor)), Ali Dixon (Co-founder (University of Bristol)), Noel Hurley (Co-founder (formerly of ARM)), Hitesh Mehta (Co-founder (Acacia Capital Partners))
- **Team size:** ~72 employees
- **Products:** XCORE, xcore.ai, VocalFusion, USB Audio solutions, GenSoC (Generative System-on-Chip)
- **Notable:** First far-field linear voice solution qualified for Amazon Alexa Voice Service (2017)., Four generations of the XCORE architecture, adding AIoT (2020) and RISC-V compatibility (2022)., Spun out the founding team of Graphcore (2016), one of the UK's most prominent AI-chip companies.

## Products & services

- **XCORE** — Multicore, deterministic microcontroller/processor architecture that runs control, I/O, DSP and AI concurrently with guaranteed real-time timing. Now spanning four generations.
- **xcore.ai** — Third-generation XCORE platform (crossover processor) built for AIoT and running neural-network inference locally at the edge, no cloud connection required.
- **VocalFusion** — Far-field voice-capture and processing solution using microphone arrays and acoustic DSP; the first far-field linear voice solution qualified for Amazon Alexa Voice Service (2017).
- **USB Audio solutions** — High-resolution and multichannel USB Audio 2.0 solutions built on XCORE for professional and consumer audio.
- **GenSoC (Generative System-on-Chip)** — 2025 recategorisation of the XCORE platform (4th gen, RISC-V compatible) enabling developers to describe system behaviour in natural language while the architecture guarantees timing and functional performance.

## Achievements

- First far-field linear voice solution qualified for Amazon Alexa Voice Service (2017).
- Four generations of the XCORE architecture, adding AIoT (2020) and RISC-V compatibility (2022).
- Spun out the founding team of Graphcore (2016), one of the UK's most prominent AI-chip companies.
- Raised roughly $107M+ total funding from investors including Amadeus, Xilinx, Infineon, Bosch and Huawei.
- Acquired SETEM Technologies (2017) to strengthen audio-algorithm capability.
- Defined the 'Generative System-on-Chip' (GenSoC) category in 2025.

## Latest updates

- **2025-10** — XMOS recategorised its fourth-generation XCORE platform as a Generative System-on-Chip (GenSoC), letting developers describe system behaviour in natural language while guaranteeing real-time timing.
- **2023-12** — Signed a joint development agreement with Sonical for Headphone 3.0 smart-headphone technology.
- **2022-12** — Announced fourth-generation XCORE with RISC-V ISA compatibility.
- **2020-02** — Launched xcore.ai, the third-generation crossover processor for AIoT with on-device neural-network inference.
- **2019-10** — Secured $19M in funding, including a venture-debt facility with Harbert European Growth Capital, to support global expansion.

## Links

- Website: https://xmos.com
- LinkedIn: https://www.linkedin.com/company/xmos
- Twitter/X: https://twitter.com/xmos
- GitHub: https://github.com/xmos
- YouTube: https://www.youtube.com/user/xmostube
- Facebook: https://www.facebook.com/XMOSMULTICORE

---

Profile page: https://yespress.io/xmos
Published by YesPress — https://yespress.io
Last updated: 2026-07-17