--- title: moonshine/micro at main · moonshine-ai/moonshine | daily.dev description: Moonshine Micro is an open source AI voice toolkit designed for resource-constrained embedded systems like microcontrollers and DSPs. It runs on the Raspberry... canonical: https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: moonshine/micro at main · moonshine-ai/moonshine | daily.dev og:description: Moonshine Micro is an open source AI voice toolkit designed for resource-constrained embedded systems like microcontrollers and DSPs. It runs on the Raspberry... og:url: https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko og:image: https://api.daily.dev/og/posts/nA15WHiKo.png og:image:alt: moonshine/micro at main · moonshine-ai/moonshine og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # moonshine/micro at main · moonshine-ai/moonshine **[Hacker News](https://daily.dev/sources/hn)** · 2 min read · 0 upvotes · 0 comments ## Summary Moonshine Micro is an open source AI voice toolkit designed for resource-constrained embedded systems like microcontrollers and DSPs. It runs on the Raspberry Pi RP2350 (80 cents) as its reference platform and fits in as little as 470 KB of RAM. The toolkit includes voice activity detection (VAD), speech-to-text (SpellingCNN), and neural text-to-speech, all powered by TensorFlow Lite Micro. The full demo pipeline uses ~3.6 MiB of flash and ~468 KiB of SRAM, with end-to-end classify-and-speak latency of 0.7–1.0 seconds. Components can be used independently and the code is MIT licensed, suitable for commercial use. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How much RAM does Moonshine Micro need to run speech-to-text and text-to-speech on a microcontroller? Moonshine Micro's demo pipeline needs about 468 KiB of provisioned SRAM total on an RP2350, since VAD, STT, and TTS run sequentially and time-share a single ~384 KiB TFLM arena rather than needing additive memory. Flash usage is around 3.6 MiB, with voice-activity detection alone using just ~36 KiB SRAM and ~89 KiB flash. _Track lightweight edge-AI toolkits like this on daily.dev before picking hardware for a voice project._ ### What hardware does Moonshine Micro use as its reference platform? Moonshine Micro uses the Raspberry Pi RP2350 microcontroller as its reference platform, a chip that retails for about 80 cents. It runs voice-activity detection, SpellingCNN-based speech-to-text, and neural diphone text-to-speech at 16 kHz, relying on TensorFlow Lite Micro for the underlying neural computations. _Developers comparing cheap microcontrollers for voice features can follow toolkits like this on daily.dev._ ## Similar posts on daily.dev - [moonshine-ai/moonshine: Fast and accurate automatic speech recognition \(ASR\) for edge devices](https://daily.dev/posts/moonshine-ai-moonshine-fast-and-accurate-automatic-speech-recognition-asr-for-edge-devices-ahlmszwb0) · Hacker News · 3 upvotes · 0 comments - [Super-Tiny AI: Running Kitten TTS \(v0.8\) on Edge Devices](https://daily.dev/posts/super-tiny-ai-running-kitten-tts-v0-8-on-edge-devices-hz45ik0ly) · SitePoint · 0 upvotes · 0 comments --- Tags: [#embedded](https://daily.dev/tags/embedded), [#tensorflow](https://daily.dev/tags/tensorflow), [#speech-recognition](https://daily.dev/tags/speech-recognition), [#text-to-speech](https://daily.dev/tags/text-to-speech) [View this post on daily.dev](https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"moonshine/micro at main · moonshine-ai/moonshine","url":"https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko"},"datePublished":"2026-07-18T22:09:26.733Z","dateModified":"2026-09-14T08:02:18.317Z","description":"Moonshine Micro is an open source AI voice toolkit designed for resource-constrained embedded systems like microcontrollers and DSPs. It runs on the Raspberry...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8cb5a3980b975ab11545583e93de98b7?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8cb5a3980b975ab11545583e93de98b7?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"embedded,tensorflow,speech-recognition,text-to-speech","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"moonshine/micro at main · moonshine-ai/moonshine"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/moonshine-micro-at-main-moonshine-ai-moonshine-na15whiko#faq","mainEntity":[{"@type":"Question","name":"How much RAM does Moonshine Micro need to run speech-to-text and text-to-speech on a microcontroller?","acceptedAnswer":{"@type":"Answer","text":"Moonshine Micro's demo pipeline needs about 468 KiB of provisioned SRAM total on an RP2350, since VAD, STT, and TTS run sequentially and time-share a single ~384 KiB TFLM arena rather than needing additive memory. Flash usage is around 3.6 MiB, with voice-activity detection alone using just ~36 KiB SRAM and ~89 KiB flash. Track lightweight edge-AI toolkits like this on daily.dev before picking hardware for a voice project."}},{"@type":"Question","name":"What hardware does Moonshine Micro use as its reference platform?","acceptedAnswer":{"@type":"Answer","text":"Moonshine Micro uses the Raspberry Pi RP2350 microcontroller as its reference platform, a chip that retails for about 80 cents. It runs voice-activity detection, SpellingCNN-based speech-to-text, and neural diphone text-to-speech at 16 kHz, relying on TensorFlow Lite Micro for the underlying neural computations. Developers comparing cheap microcontrollers for voice features can follow toolkits like this on daily.dev."}}]} ```