Top all-time · #3090
On-device AI for real-time Voice, Language, and Vision Understanding
Company blog · picovoice.ai/blog · on HN as kenarsaaiprogramming
On-device AI for real-time Voice, Language, and Vision Understanding is the company blog at picovoice.ai, writing about ai and programming. Since 2018, 25 different Hacker News readers have linked it 62 times, most often “Voice control engine that runs fully in the browser using WebAssembly”. No feed was found, so read it on the site itself.
Too few sampled comments (19) to say how often HN recommends it.?
Why HN links it
HN readers link Picovoice's product pages, docs, FAQ, pricing and careers pages for its on-device voice AI tools such as wake word and speech-to-text.
Summarised by Claude from a sample of 19 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Production-grade on-device AI SDKs for voice, language, and vision. STT, wake word, TTS, translation, OCR, and more. Runs on mobile, web, desktop, and embedded.
Most-linked posts?
- /pricing/4 people · 5 mentions · first linked 2022-11-22
- Voice control engine that runs fully in the browser using WebAssembly3 people · 3 mentions · first linked 2018-04-02
- /platform/cat/2 people · 3 mentions · first linked 2022-05-19
- Porcupine Wake Word Detection Engine2 people · 2 mentions · first linked 2023-07-03
- /platform/cheetah/2 people · 2 mentions · first linked 2023-05-15
- Picovoice – Embed private voice AI into any product instantly2 people · 2 mentions · first linked 2018-10-12
What Hacker News says?
You can check out PicoVoice Eagle (paid product): https://picovoice.ai/docs/eagle/ You pass N number of PCM frames through their trainer and once you reach a certain percentage you can extract an embedding you can save. Then you can identify audio against the set of identified speakers and it will return percentage matches for each.
Comment linking /docs/eagle/
on “OpenVoice: Versatile instant voice cloning”Submitted Picovoice – Embed private voice AI into any product instantly as a story · 182 points
HN discussionoh no, i meant mark II, the speaker. I use their picovoice sdk. it has wake word and intent detectors - Porcupine and Rhino. https://picovoice.ai/docs/picovoice/ and https://ttsmp3.com/ i was going to add picovoice's speech to text with rasa https://rasa.com/ but i didnt have time, will give it a try over the holidays. I see your point but not every open source project lives forever, like sonos killed snips.
Comment linking /docs/picovoice/
on “Mycroft – open source voice assistant”Submitted Voice control engine that runs fully in the browser using WebAssembly as a story · 86 points
HN discussionI gave up on mycroft after a long wait and built my own with respeaker and picovoice. i have 2 of them with different wake words. imo it's way better and easier than snowboy. i dont understand why people give their data to amazon to set a timer :) check their free stuff: https://picovoice.ai/pricing/
Comment linking /pricing/
on “Mycroft – open source voice assistant”Submitted Towards Optimal LLM Quantization as a story · 18 points
HN discussionDeep Learning Researcher - Vancouver - Onsite https://picovoice.ai/careers/deep-learning-researcher/ Picovoice is the only all-in-one on-device voice AI platform with offerings including wake word, speech-to-text, LLM, text-to-speech, and more. All on-device. Runs across mobile, web, desktop, and embedded.
Comment linking /careers/deep-learning-researcher/
on “Ask HN: Who is hiring? (October 2025)”Submitted Offline Voice AI in a Web Browser as a story · 12 points
HN discussionI saw a whole PR release for adtech oriented keyword spotting tech like this one time, I wish I'd saved it... I can't remember the name of that particular company. But read through this doc: https://picovoice.ai/docs/faq/porcupine/
Comment linking /docs/faq/porcupine/
on “I still don't think companies serve you ads based on your microphone”Submitted How to Record Audio from a Web Browser as a story · 6 points
HN discussionFor speech-to-text right? I provided the quote for voice control engine: https://picovoice.ai/#wake-word-detection The voice control comes in two variations standard and tiny. The tiny one consumes even fewer resources. I provided metrics for the standard one. You can check the benchmark repo on benchmark it yourself as well :) https://github.com/Picovoice/wakeword-benchmark
Comment linking /
on “Picovoice – Embed private voice AI into any product instantly”Submitted Voice-Enabled Elevator as a story · 6 points
HN discussionI am the founder of Picovoice. I am NOT going to comment on comparison as obviously I will be biased ;) With Picovoice you can use the voice control engine to accomplish this. Maybe something similar to this demo? https://picovoice.ai/#voice-control-demo The cool thing about this engine is that it is tiny. It uses less than 8% CPU on RPi3 and altogether it is less than 2MB (code, model, etc). Technically you can run…
Comment linking /
on “Picovoice – Embed private voice AI into any product instantly”Submitted Offline Voice Assistant on a Microcontroller with 192KB RAM as a story · 5 points
HN discussionAssuming the hardware kill switches work, you're right that the wake word is a big issue. Would be nice if it worked offline with something like this: https://picovoice.ai/products/porcupine.html A demo of it working without any internet would be cool (even though calls require internet)
Comment linking /products/porcupine.html
on “Facebook launches Portal TV, a $149 video chat set-top box”Submitted Porcupine Wake Word Detection Engine as a story · 4 points
HN discussionhttps://picovoice.ai/careers/applied-speech-scientist/ Picovoice is a deep tech startup founded by engineers and driven by engineers. We are accelerating the transition of voice AI from the cloud to the edge. Why? Privacy, reliability, and the environment. Numerous enterprises, including NASA and Stanford University, are leveraging Picovoice technology in their products. In this role, you get to work on Applied rese…
Comment linking /blog/direct-speech-indexing/
on “Ask HN: Who is hiring? (December 2022)”Submitted How to Improve Speech-to-Text Accuracy as a story · 4 points
HN discussionfirst, good initiative! thanks for sharing. i think you gotta be more diligent and careful with the problem statement. checking the weather in Sofia, Bulgaria requires cloud, current information. it's not "random speech". ESP SR capability issues don't mean that you cannot process it locally. the comment was on "voice processing" i.e. sending speech to the cloud, not sending a call request to get the weather informa…
Comment linking /platform/cheetah/
on “Show HN: Willow – Open-Source Privacy-Focused Voice Assistant Hardware”Submitted Private voice recognition running locally in browser as a story · 4 points
HN discussionLooks more like an ad for https://picovoice.ai/products/cheetah.html
Comment linking /products/cheetah.html
on “Offline Speech-to-Text on Raspberry Pi Zero”Submitted NLU on Arduino as a story · 3 points
HN discussionPicovoice runs on almost anything: web browsers, mobile, desktop, single board computers, and microcontrollers. For the platforms that have connectivity (i.e. almost anything aside from microcontrollers), we do call home for license management. This helps us keep the `Free Tier` free for personal users, hackers , and skunkworks projects, but make sure we get paid by enterprise customers with deployments at scale [1]…
Comment linking /pricing/
on “Offline Voice Assistant on a Microcontroller with 192KB RAM”Submitted Speak Privately to the Browser as a story · 3 points
HN discussioncontext, context, and context!! Then deep learning tricks and efficient implementation. [1] https://picovoice.ai/blog/end-to-end-intent-inference-from-s...
Comment linking /blog/end-to-end-intent-inference-from-speech/
on “Offline Voice Assistant on a Microcontroller with 192KB RAM”Submitted Cops Taking Your Voice as a story · 2 points
HN discussionhave you noted that there is `Free Tier` that cost you $0? You can train using that. For this tutorial the cost of board ($20) is all you need to pay. Same for personal projects and even small skunkworks projects within companies. Picovoice makes money from large-scale deployments done by device makers [1] https://picovoice.ai/pricing/
Comment linking /pricing/
on “Offline Voice Assistant on a Microcontroller with 192KB RAM”Submitted Tutorial: Speech-to-Text with Node.js as a story · 2 points
HN discussionFrom their pricing FAQ[1]: 19. Why do Picovoice engines require an internet connection? While data is processed offline, locally on-device, Picovoice engines call home servers to stay active and report the consumption for billing purposes only. [1] https://picovoice.ai/pricing/
Comment linking /pricing/
on “Offline Voice Assistant on a Microcontroller with 192KB RAM”Submitted Polyglot Voice Interfaces at the Edge as a story · 2 points
HN discussion