Xiaozhi Linux
Open-source port of the Xiaozhi AI voice chatbot to embedded Linux single-board computers
Choose Xiaozhi Linux if you are an embedded developer who wants the Xiaozhi voice assistant running on a Linux SBC and are willing to port board support yourself — the verified support list covers i.MX6ULL, Allwinner T113/V85x, K230, Rockchip, and STM32MP157. The architecture is the selling point: AI logic is kept separate from board BSP and drivers, so a new board is a contained job rather than a fork. Skip it if you want a finished product, or if you need English documentation: architecture pages, community (QQ group 564992486), and the optional AIoT edge AI course are all Chinese-language. For mature open-source voice wake-word tooling on embedded Linux, look at Picovoice's Porcupine and
Verified 5d ago · liveness 56/100 · cite: rightaichoice.com/tools/xiaozhi-linux
- Embedded Linux developers building voice assistants on SBCs
- Makers integrating voice interaction into custom hardware
- Hardware engineers porting AI voice logic to new boards
- Chinese-speaking IoT and AIoT developers
- Buyers who want a plug-and-play consumer voice device
- Absolute beginners without embedded Linux or cross-compilation experience
- Teams that need English-language documentation and community
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Xiaozhi Linux if you want a finished product rather than a board-port project, or if you need English architecture docs and a paid support contract instead of a QQ group and GitHub issues.
A supported board is not included — you buy the SBC (for example 100ASK_T113s3-PRO or 100ASK_IMX6ULL-PRO) separately before you can run anything.
The Linux AI Xiaozhi core code is published on GitHub as an open-source project, so the software cost is your engineering time; the real spend is the SBC itself and, optionally, the paid AIoT edge AI course. Compared with hosted voice-assistant services that bill per request, this model shifts cost from subscriptions to hardware and developer hours, which favours teams already comfortable with embedded Linux.
In short
Xiaozhi Linux — Open-source port of the Xiaozhi AI voice chatbot to embedded Linux single-board computers. Best for Embedded Linux developers building voice assistants on SBCs, Makers integrating voice interaction into custom hardware, Hardware engineers porting AI voice logic to new boards. Free to use.
What people actually say about Xiaozhi Linux — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
8 mentions across 1 source (GitHub) · researched Jul 3, 2026.
Average across the 1 source that answered — each source counts once, not each post.
- +Free and open-source under MIT-like license.
- +Runs on low-cost embedded Linux boards without cloud dependencies.
- +Separation of core AI logic from board-specific BSP/drivers simplifies porting.
- +Supports multiple popular SBC families: NXP, Allwinner, Rockchip, STM32.
- +Voice AI with natural language processing and speech output on resource-constrained hardware.
- −Development appears stalled with no recent updates.
- −Documentation and community are entirely in Chinese.
- −Compilation often fails due to missing dependencies.
- −Setup links for supported boards give 404 errors.
- −No official support for non-Chinese users.
- • Hardware cost of supported SBCs and peripherals.
- • Time cost for debugging compilation errors and broken links.
Viability Score
How well maintained and how widely used is Xiaozhi Linux? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Voice input and speech output on embedded Linux SBCs
- Ports the AI Xiaozhi chatbot design from ESP32 to Linux
- Supported hardware: NXP i.MX6ULL, Allwinner T113/V85x, Canaan K230, Rockchip series, STM32MP157
- Core AI logic kept separate from board BSP drivers and dependencies
- Verified board list including DshanPI-CanMV K230, 100ASK_T113s3-PRO (Tina4/Tina5), DongshanPI-AICT, 100ASK_IMX6ULL-PRO
- sound_app framework for the audio pipeline
- control_center framework for orchestration
- Documented Xiaozhi interface flow architecture
- xioazhi-linux-SupportList repository for submitting new board ports
- Community issue and pull request workflow on GitHub
- Chinese-language architecture documentation
- Community support via QQ group 564992486
- Bilibili run demonstration video
- Optional paid AIoT edge AI large-model course
- Open-source licensing as published in the repository
About Xiaozhi Linux
Xiaozhi Linux is an open-source project that brings the Xiaozhi AI voice chatbot to embedded Linux single-board computers. It ports the design of the ESP32 AI Xiaozhi project to Linux SBCs including the NXP i.MX6ULL, Allwinner Tina T113 and V85x series, Canaan K230, Rockchip series, and STM32MP157. The core repository holds only the AI logic while board-specific BSP drivers and dependencies stay separate, so adapting the assistant to new hardware means adding a board port rather than rewriting the AI stack. The project ships with a support list of verified boards, a demonstration video on Bilibili, Chinese architecture docs covering the sound_app framework, the control_center framework, and the Xiaozhi interface flow. It is aimed at embedded Linux developers and hardware makers who want voice interaction running on their own board — not a consumer appliance — with the trade-off that you need real embedded Linux experience and that documentation and community support are Chinese-language, delivered through a QQ group (564992486) and GitHub issues and pull requests.
Behind the Verdict
Xiaozhi Linux solves a specific problem well: it takes the ESP32 AI Xiaozhi chatbot design and makes the AI logic run on embedded Linux boards. The repository deliberately holds only the Linux AI Xiaozhi core code and excludes per-board hardware BSP drivers and dependencies, which is why the supported-hardware list reads as ports rather than forks. That separation is the strongest engineering decision here — you can add a board by submitting support to the xiaozhi-linux-SupportList repository, and the project explicitly invites issues and pull requests to improve it. The documentation set covers the architecture in Chinese: an overview of the Xiaozhi interface flow, an analysis of the sound_app framework, and an analysis of the control_center framework. There is also a Bilibili run demo you can watch before committing to a board. Community help routes through a QQ group (564992486) and GitHub, and there is an optional paid course on AIoT edge AI large-model development if you want structured learning. Where it does not fit: this is not a product for someone who wants to unbox a voice assistant. It assumes embedded Linux competence — buildroot or Tina images, cross-compilation, and board bring-up are part of the job. Documentation and support are Chinese-first, which raises the effort for non-Chinese-speaking developers. And because the repository excludes board BSPs, you should expect a non-trivial integration step even on a supported board. Bottom line: a credible, actively developed community project for hardware developers who want voice AI on their own SBC and care about controlling the whole stack, rather than for teams that need commercial support or a turnkey device.
Researching Xiaozhi Linux? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Xiaozhi Linux actually fits — and what changes day-one when you adopt it.
Clones the xiaozhi-linux repository to get the AI 小智 core logic, matches it against the 100ASK_IMX6ULL-PRO Buildroot20 entry on the supported-hardware list, adds the board BSP from the corresponding platform repository, then cross-compiles and boots the assistant to verify voice input and speech output.
Outcome: A working Xiaozhi voice assistant on their own i.MX6ULL hardware, with the AI logic kept apart from board drivers so future board changes don't touch the core.
Picks the 100ASK_T113s3-PRO Tina4 or Tina5 verified board entry, follows the Chinese architecture docs on the Xiaozhi interface flow, the sound_app framework, and the control_center framework, and uses the Bilibili demo video as the reference for expected behaviour.
Outcome: A functioning voice interaction prototype on T113 hardware, reached without reverse-engineering the audio pipeline from scratch.
Studies the sound_app and control_center frameworks, then submits a board port through issues and pull requests to the xiaozhi-linux-SupportList repository so the new board joins the supported list.
Outcome: Their board becomes a documented, community-supported target for the Linux Xiaozhi project.
Use Cases
- Run an offline-capable voice assistant on your own NXP i.MX6ULL board for a home automation setup.
- Build a voice-driven kiosk on Allwinner T113 hardware for a retail or industrial installation.
- Add voice control to a robotics project running on a Rockchip SBC.
- Study the sound_app and control_center frameworks to learn how the Xiaozhi assistant is structured on Linux.
- Submit a board port to the xiaozhi-linux-SupportList repository to add support for a new SBC.
Limitations
- The repository deliberately contains only the Linux AI Xiaozhi core code — board-specific hardware BSP drivers and dependencies live elsewhere, so getting it running on any board is an integration job rather than a flash-and-go.
- Architecture docs, the QQ community, and the optional AIoT course are all Chinese-language, which adds friction for non-Chinese-speaking developers.
- Support is community-driven through QQ group 564992486, GitHub issues, and pull requests, so response times depend on volunteers rather than a support contract.
- The verified hardware list is finite — i.MX6ULL, Allwinner T113/V85x, K230, Rockchip, STM32MP157 and specific named boards — and anything else requires you to submit a port to the SupportList repository.
as of 2026-10-03
Verification history
We have re-verified Xiaozhi Linux 9 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
- — re-checked, vendor evidence unchanged
Showing the 6 most recent of 9 verification passes.
Free to cite with attribution — this page re-verifies continuously.
Where the pricing makes sense
The company stage and team size where Xiaozhi Linux's pricing actually pencils out — and where peers do it cheaper.
The Linux AI Xiaozhi core code is published on GitHub as an open-source project, so the software cost is your engineering time; the real spend is the SBC itself and, optionally, the paid AIoT edge AI course. Compared with hosted voice-assistant services that bill per request, this model shifts cost from subscriptions to hardware and developer hours, which favours teams already comfortable with embedded Linux.
Setup time & first value
How long it actually takes to get something useful out of Xiaozhi Linux — broken out by persona, not the marketing-page minute.
For a developer already fluent in embedded Linux and cross-compilation on an officially listed board such as 100ASK_IMX6ULL-PRO or 100ASK_T113s3-PRO, expect an afternoon to a day to get the core running once the board BSP is in place. For an unlisted board, budget days of bring-up work before the AI logic ever executes, since you are writing the port as you go.
Switching to or from Xiaozhi Linux
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From ESP32 AI Xiaozhi: move the assistant to Linux by using the xiaozhi-linux core, which was built by porting the ESP32 design, and supply your board's BSP separately.
- →From a cloud voice assistant service: replace the hosted endpoint with local Xiaozhi logic running on your own SBC, at the cost of doing board bring-up yourself.
- →From a proprietary voice SDK on Linux: swap in the open-source Xiaozhi core and follow the sound_app and control_center documentation to reproduce the audio pipeline.
- ↗To a commercial embedded voice SDK: move off Xiaozhi Linux if you need vendor support contracts and English documentation rather than QQ-group and GitHub help.
- ↗To ESP32 AI Xiaozhi: go back to the ESP32 design if your target hardware is a microcontroller rather than a Linux SBC.
- ↗To a hosted voice assistant service: move off if you would rather rent inference per request than maintain board ports in-house.
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “Xiaozhi Linux”, and we withheld 6: 6 did not mention Xiaozhi Linux. We are showing none, because we could not prove any of them are about Xiaozhi Linux.
Official links
Tools that pair well with Xiaozhi Linux
Common stack mates teams adopt alongside Xiaozhi Linux, with the specific reason each pairing earns its keep.
ElevenLabs Conversational AI
Deploy low-latency conversational AI voice agents across 90+ languages in minutes.
Hey Internet
Heynet puts AI employees — Sales, Marketing, Research and Executive Assistant agents — to work on account research, outreach and content.
Lightberry
Lightberry is the social brain layer for the Lumi, a 1.3-meter humanoid robot you talk to, that sees its surroundings and learns from the people around it.
Featured Head-to-Head Comparisons
Xiaozhi Linux vs Spider Cloud
If you're building an AI voice assistant on embedded Linux, Xiaozhi Linux is a free, open-source choice. For web data extraction to feed AI agents or RAG pipelines, Spider Cloud's fast Rust engine and rich integrations are far more appropriate. These tools serve entirely different purposes, so your decision hinges on your project domain.
Xiaozhi Linux vs Temporal Ai
Xiaozhi Linux is the go-to for embedded voice AI on low-power SBCs, while Temporal AI dominates durable execution for cloud-native workflows. Choose Xiaozhi if you need offline-capable voice on i.MX6ULL or STM32; pick Temporal for building crash-resistant AI agents and distributed workflows with retries. They serve entirely different needs and are complementary rather than competing.
Xiaozhi Linux vs Voyage Ai
Choose Voyage AI if you need premium embedding/reranker models for enterprise RAG with domain specialization and compliance. Choose Xiaozhi Linux if you are building an offline-capable voice assistant on embedded Linux SBCs and value open-source flexibility. They serve completely different markets — Voyage for cloud-based NLP retrieval, Xiaozhi for edge voice interaction.
Alternatives to Xiaozhi Linux
View all
ElevenLabs Conversational AI
Deploy low-latency conversational AI voice agents across 90+ languages in minutes.
Hey Internet
Heynet puts AI employees — Sales, Marketing, Research and Executive Assistant agents — to work on account research, outreach and content.
Lightberry
Lightberry is the social brain layer for the Lumi, a 1.3-meter humanoid robot you talk to, that sees its surroundings and learns from the people around it.
Frequently Asked Questions
Best-of guides
Topics
Used Xiaozhi Linux? Help shape our editorial sentiment research.