_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
HTML Visit Hacker News on the Web
COMMENT PAGE FOR:
HTML Show HN: Moonshine Open-Weights STT models â higher accuracy than WhisperLargev3
Ross00781 wrote 4 hours 13 min ago:
Open-weight STT models hitting production-grade accuracy is huge for
privacy-sensitive deployments. Whisper was already impressive, but
having competitive alternatives means we're not locked into a single
model family. The real test will be multilingual performance and edge
device efficiencyâhas anyone benchmarked this on M-series or Jetson?
fudged71 wrote 4 hours 59 min ago:
If it's using ONNX, can this be ported to Transformers.js?
fittingopposite wrote 9 hours 49 min ago:
Which program does support it to allow streaming? Currently using
spokenly and parakeet but would like to transition to a model that is
streaming instead of transcribing chunk wise.
regularfry wrote 11 hours 20 min ago:
Oh this is fantastic. I'm most interested to see if this reaches down
to the raspberry pi zero 2, because that's a whole new ballgame if it
does.
dSebastien wrote 11 hours 58 min ago:
I've been using Moonshine since V1 and the results are really great.
I'd say on par with Parakeet V3 while working really well with CPU
only.
T0mSIlver wrote 13 hours 11 min ago:
Congrats on the results. The streaming aspect is what I find most
exciting here.
I built a macOS dictation app ( [1] ) on top of Voxtral Realtime, and
the UX difference between streaming and offline STT is night and day.
Words appearing while you're still talking completely changes the
feedback loop. You catch errors in real time, you can adjust what
you're saying mid-sentence, and the whole thing feels more natural.
Going back to "record then wait" feels broken after that.
Curious how Moonshine's streaming latency compares in practice. Do you
have numbers on time-to-first-token for the streaming mode? And on the
serving side, do any of the integration options expose an OpenAI
Realtime-compatible WebSocket endpoint?
HTML [1]: https://github.com/T0mSIlver/localvoxtral
sourcetms wrote 14 hours 53 min ago:
I'm offering support for this in Resonant - Already set up and running
this week.
It's incredible for a live transcription stream - the latency is WOW.
[1] For the open source folks, that's also set up in handy, I think.
HTML [1]: https://www.onresonant.com/
admiralrohan wrote 13 hours 44 min ago:
Is this alternative to Whispr Flow?
binome wrote 15 hours 30 min ago:
I vibe-trained moonshine-tiny on amateur radio morse code last weekend,
and was surprised at the ~2% CER I was seeing in evals and over the air
performance was pretty acceptable for a couple hour run on a 4090.
Ross00781 wrote 16 hours 15 min ago:
The streaming architecture looks really promising for edge deployments.
One thing I'm curious about: how does the caching mechanism handle
multiple concurrent audio streams? For example, in a meeting
transcription scenario with 4-5 speakers, would each stream maintain
its own cache, or is there shared state that could create bottlenecks?
dagss wrote 17 hours 7 min ago:
Very exciting stuff!
hear about what people might build with it
My startup is making software for firefighters to use during missions
on tablets, excited to see (when I get the time) if we can use this as
a keyboard alternative on the device. It's a use case where avoiding
"clunky" is important and a perfect usecase for speech-to-text.
Due to the sector being increasingly worried about "hybrid threats" we
try to rely on the cloud as little as possible and run things either on
device or with the possibility of being self-hosted/on-premise. I
really like the direction your company is going in in this respect.
We'd probably need custom training -- we need Norwegian, and there's
some lingo, e.g., "bravo one two" should become "B-1.2". While that can
perhaps also be done with simple post-processing rules, we would also
probably want such examples in training for improved recognition? Have
no VC funding, but looking forward to getting some income so that we
can send some of it in your direction :)
steinvakt2 wrote 13 hours 25 min ago:
Interesting. Can we get in touch? I just sold my webapp/saas where I
used NB-Whisper to transcribe Norwegian media (podcast, radio, TV)
and offer alerts and search by indexing it using elasticsearch.
Edit: It was [1] (I shut down the backend server yesterday so the
functionality is disabled).
HTML [1]: https://muninai.eu
dagss wrote 12 hours 12 min ago:
Sure! I didn't find your contact info but drop me an email at
dag@syncmap.no.
RobotToaster wrote 17 hours 12 min ago:
> Models for other languages are released under the Moonshine Community
License, which is a non-commercial license.
Weird to only release English as open weights.
riedel wrote 15 hours 40 min ago:
I find it an even more weird practice for anyone working with speech
or text models not in the first paragraph name the language it is
meant for (and I do not mean the programming language bindings). How
many English native speakers are there 5% of the world population?
RobotToaster wrote 14 hours 55 min ago:
Approximately yes, although another 15% are non-native English
speakers. Chinese is a close second for total speakers.
raybb wrote 19 hours 3 min ago:
fyi the typepad link in your bio is broken
oezi wrote 19 hours 31 min ago:
Do you also support timestamps the detected word or even down to
characters?
guerython wrote 19 hours 44 min ago:
Nice work. One metric Iâd really like to see for streaming use cases
is partial stability, not just final WER.
For voice agents, the painful failure mode is partials getting
rewritten every few hundred ms. If you can share it, metrics like
median first-token latency, real-time factor, and "% partial tokens
revised after 1s / 3s" on noisy far-field audio would make comparisons
much more actionable.
If those numbers look good, this seems very promising for local
assistant pipelines.
regularfry wrote 9 hours 9 min ago:
Tangentially, have you got any idea what the equivalent "partial
tokens revised" rate for humans is? I know I've consciously
experienced backtracking and re-interpreting words before, and
presumably it happens subconsciously all the time. But that means
there's a bound on how low it's reasonable to expect that rate to be,
and I don't have an intuition for what it is.
heftykoo wrote 19 hours 57 min ago:
Claiming higher accuracy than Whisper Large v3 is a bold opening move.
Does your evaluation account for Whisper's notorious hallucination
loops during silences (the classic 'Thank you for watching!'), or is
this purely based on WER on clean datasets? Also, what's the VRAM
footprint for edge deployments? If it fits on a standard 8GB Mac
without quantization tricks, this is huge.
starkparker wrote 20 hours 13 min ago:
Implemented this to transcribe voice chat in a project and the
streaming accuracy in English on this was unusable, even with the
medium streaming model.
francislavoie wrote 21 hours 19 min ago:
I've helped many Twitch streamers set up [1] to plug transcription &
translation into their streams, mainly for German audio to English
subtitles.
I'd love a faster and more accurate option than Whisper, but streamers
need something off-the-shelf they can install in their pipeline, like
an OBS plugin which can just grab the audio from their OBS audio
sources.
I see a couple obvious problems: this doesn't seem to support
translation which is unfortunate, that's pretty key for this usecase.
Also it only supports one language at a time, which is problematic with
how streamers will frequently code-switch while talking to their chat
in different languages or on Discord with their gameplay partners.
Maybe such a plugin would be able to detect which language is spoken
and route to one or the other model as needed?
HTML [1]: https://github.com/royshil/obs-localvocal
saltwounds wrote 21 hours 30 min ago:
Streaming transcription is crazy fast on an M1. Would be great to use
this as a local option versus Wispr Flow.
fareesh wrote 22 hours 16 min ago:
Accuracy is often presumed to be english, which is fine, but it's a
vague thing to say "higher" because does it mean higher in English
only? Higher in some subset of languages? Which ones?
The minimum useful data for this stuff is a small table of language |
WER for dataset
999900000999 wrote 22 hours 18 min ago:
Very cool. Anyway to run this in Web assembly, I have a project in mind
alexnewman wrote 22 hours 19 min ago:
If only it did Doric
nmstoker wrote 22 hours 20 min ago:
Any plans regarding JavaScript support in the browser?
There was an issue with a demo but it's missing now. I can't recall for
sure but I think I got it working locally myself too but then found it
broke unexpectedly and I didn't manage to find out why.
Karrot_Kream wrote 23 hours 2 min ago:
According to the OpenASR Leaderboard [1], looks like Parakeet V2/V3 and
Canary-Qwen (a Qwen finetune) handily beat Moonshine. All 3 models are
open, but Parakeet is the smallest of the 3. I use Parakeet V3 with
Handy and it works great locally for me.
[1]
HTML [1]: https://huggingface.co/spaces/hf-audio/open_asr_leaderboard
Imustaskforhelp wrote 9 hours 28 min ago:
To this comment and all the other comments talking about handy below
this comment. I tried handy right now and it's super amazing. I'm
speaking this from Handy. This is so cool, man.
And handy even takes care of all the punctuation, which is really
nice.
Thanks a lot for suggesting it to me. I actually wanted something
like this, and I was using something like Google Docs, and it
required me to use Chrome to get the speech to text version, and I
actually ended up using Orion for that because Orion can actually
work as a Chrome for some reason while still having both Firefox and
Chrome extension support. So and I had it installed, but yeah.
This is really amazing and actually a sort of lifesaver actually, so
thanks a lot, man.
Now I can actually just speak and this can convert this to text
without having to go through any non-local model or Google Docs or
whatever anything else.
Why is this so good man? It's so good
man, I actually now am thinking that I had like fully maxed out my
typing speed to like hundred-120. But like this can actually write it
faster. you know it's pretty amazing actually.
Have a nice day, or as I abbreviate it, HAND, smiley face. :D
d4rkp4ttern wrote 10 hours 25 min ago:
Was a big fan of Handy until I found Hex, which, incredibly, has even
faster transcription (with Parakeet V3), itâs MacOS only:
HTML [1]: https://github.com/kitlangton/Hex
Imustaskforhelp wrote 9 hours 55 min ago:
I tried this out but the brew command errors out saying it only
works on macOS versions older than Sequoia.
That's unfortunate. I think I can update my version but I have
heard some bad things about performance from the newer update from
my elder brother.
ValentineC wrote 4 hours 55 min ago:
> I tried this out but the brew command errors out saying it only
works on macOS versions older than Sequoia.
Newer than Sequoia, you mean?
The brew recipe [1] says macOS >= 15.
Anyway, I'm on Sequoia â it's mostly better than Ventura, which
was what my M2 MacBook Pro came with. I'm holding off upgrading
to Tahoe (macOS 26), hoping they fix liquid glAss.
HTML [1]: https://formulae.brew.sh/cask/kitlangton-hex
d4rkp4ttern wrote 9 hours 6 min ago:
works fine on my MacOS w Tahoe
kardaj wrote 12 hours 21 min ago:
I'm building a local-first transcription iOS app and have been on
Whisper Medium, switching to Parakeet V3 based on this.
One note for anyone using Handy with codex-cli on macOS: the default
"Option + Space" shortcut inserts spaces mid-speech. "Left Ctrl + Fn"
works cleanly instead. I'm curious to know which shortcuts you're
using.
bn-usd-mistake wrote 11 hours 55 min ago:
I am looking for such an app. Main use case is transcribing voice
notes received on Signal while preserving privacy. Please post when
you launch :)
agentifysh wrote 18 hours 43 min ago:
hmmm looks like assembyAI is still unbeatable here in terms of
cost/performance unless im mistaken
edit: holy shit parakeet is good.... Moonshine impressive too and it
is half the param
Now if only there was something just as quick as Parakeet v3 for TTS
! Then I can talk to codex all day long!!!
fittingopposite wrote 9 hours 55 min ago:
Also running parakeet on my phone with [1] Very lightweight and
good quality
HTML [1]: https://github.com/notune/android_transcribe_app
agentifysh wrote 1 hour 46 min ago:
This is actually pretty impressive. What kinda phone are you
using? Are you noticing any drain on battery heat?Do you think
it's possible to get this working with Flutter on iOS?
fittingopposite wrote 47 min ago:
2-3 years old Android flagship phone with 8 GB RAM. When I
looked for an app for parakeet, I think I also came across iOS
apps. Don't recall it since I use Android.
Seems light on the phone/battery. Don't observe any drain but I
also only record shorter transcripts at once.
Side note: Parakeet is actually pretty nice to do meetings with
oneself. Did that on a computer while driving for an hour
(split in several transcript chunks). Processed the raw meeting
notes afterwards with an LLM. Effective use of the time in the
car...
agentifysh wrote 38 min ago:
Thank you for sharing !
What about the quality of the transcripts? Is it able to do
live streaming?
fittingopposite wrote 30 min ago:
Unfortunately, Parakeet doesn't support streaming like
Moonshot does (as much as I know). Would be perfect to have
sth of the size of Parakeet but supporting streaming. Still
hope Nvidia releases a V4 with that feature :)
Otherwise, I think STT is basically a solved problem
running locally on edge devices.
Dayshine wrote 15 hours 45 min ago:
What's wrong with piper?
remuskaos wrote 18 hours 10 min ago:
Parakeet doesn't require a GPU. I'm handily running it on my Ubuntu
Linux laptop.
namibj wrote 14 hours 43 min ago:
I'm looking to switch from feeding the default android "recorder"
app's .WAV into Gemini 3 Pro (via the app) with (usually just) a
`Transcribe this please:` prompt; content is usually German voice
instructions/explanation for how to do/approach some sysadmin
stuff; there does tend to be some amount of interjecting
(primarily for clarifications(-posing/-requesting)) by me to
resolve ambiguity as early as possible/practical.
If e.g. parakeet can be run on my phone in real time showing the
transcript live:
- with latency low enough to be "comfortable enough" for the
instructor to keep an eye on and approve the transcribed
instructions
[not necessarily every word of the transcript, i.e., a commanded
"edit" doesn't need to be applied in the outcome as long as it's
nature is otherwise clear enough to not add meaningful amounts of
ambiguity to the final "written" instructions]
by glancing at the screen while dictating the explanation (and
blurting out any transcription complaints as soon as that's
possible without breaking one's own string-of-thought or spoken
grammar too much)
, I'd very happily switch to that approach instead of what I was
doing.
Bonus if there's a no-bulky-or-expensive-hardware way to
accommodate us both speaking over each other so I won't have to
_interrupt_ his speaking just to put a clarifying comment (on
what he just said) in the transcript for him to see and sign off,
where the at least "only" briefly interrupts his thoughts right
while he actually reads my transcribed words (he doesn't have to
hear them, and it's better if he won't; I can probably get him to
put on earmuffs to not hear me louder than he hears his thoughts,
and a sufficiently-smoothed SNR meter for specifically his voice
should take care him regulating his volume while the earmuffs
mute it and I occasionally talk over him)...
agentifysh wrote 18 hours 7 min ago:
you are right i just downloaded it on handy and its working i
can't believe it
i was using assmeblyAI but this is fast and accurate and offline
wtf!
tuananh wrote 19 hours 32 min ago:
Handy is amazing. Super quality app.
agentifysh wrote 17 hours 26 min ago:
It really is. It's kinda ridiculous that it's free.
tuananh wrote 14 hours 22 min ago:
I'm quite surprise to see that level of polish from an
open-source project.
alfiedotwtf wrote 14 hours 55 min ago:
Are voice or a transcript sent back to their servers? If so, you
may be the product
yorwba wrote 14 hours 42 min ago:
No, it's just somebody's open source project:
HTML [1]: https://github.com/cjpais/handy
tomr75 wrote 20 hours 36 min ago:
why V3 over V2 (assuming English only)?
syntaxing wrote 20 hours 52 min ago:
How much VRAM does parakeet take for you? For some reason it takes
4GB+ for me using the onyx version even though itâs 600M parameters
theologic wrote 21 hours 10 min ago:
By the way, I've been using a Whisper model, specifically WhisperX,
to do all my work, and for whatever reason I just simply was not
familiar with the Handy app. I've now downloaded and used it, and
what a great suggestion. Thank you for putting it here, along with
the direct link to the leaderboard.
I can tell that this is now definitely going to be my go-to model and
app on all my clients.
jasonjmcghee wrote 18 hours 10 min ago:
I have to ask- I see this handy app running on Mac and you hold a
key down and then it doesn't show until seemingly a while later.
The one built in is much faster, and you only have to toggle it on.
Are these so much more accurate? I definitely have to correct
stuff, but pretty good experience.
Also use speech to text on my iphone which seems to be the same
accuracy.
reitzensteinm wrote 21 hours 41 min ago:
Parakeet V3 is over twice the parameter count of Moonshine Medium
(600m vs 245m), so it's not an apples to apples comparison.
I'm actually a little surprised they haven't added model size to that
chart.
bytesandbits wrote 16 hours 24 min ago:
parakeet v3 has a much better RTFx than moonshine, it's not just
about parameter numbers. Runs faster.
HTML [1]: https://huggingface.co/spaces/hf-audio/open_asr_leaderboar...
SyneRyder wrote 4 hours 8 min ago:
That was my experience when I tried Moonshine against Parakeet v3
via Handy. Moonshine was noticeably slower on my 2018-era Intel
i7 PC, and didn't seem as accurate either. I'm glad it exists,
and I like the smaller size on disk (and presumably RAM too). But
for my purposes with Handy I think I need the extra speed and
accuracy Parakeet v3 is giving me.
regularfry wrote 7 hours 33 min ago:
It is about the parameter numbers if what you care about is edge
devices with limited RAM. Beyond a certain size your model just
doesn't fit, it doesn't matter how good it is - you still can't
run it.
agentifysh wrote 17 hours 24 min ago:
So I'm kinda new to this whole parakeet and moonshine stuff, and
I'm able to run parakeet on a low end CPU without issues, so I'm
curious as to how much that extra savings on parameters is actually
gonna translate.
Oh and I type this in handy with just my voice and parakeet version
three, which is absolutely crazy.
asqueella wrote 23 hours 6 min ago:
For those wondering about the language support, currently English,
Arabic, Japanese, Korean, Mandarin, Spanish, Ukrainian, Vietnamese are
available (most in Base size = 58M params)
sroussey wrote 23 hours 10 min ago:
onnx models for browser possible?
pzo wrote 23 hours 16 min ago:
haven't tested yet but I'm wondering how it will behave when talking
about many IT jargon and tech acronyms. For those reason I had to
mostly run LLM after STT but that was slowing done parakeet inference.
Otherwise had problems to detect properly sometimes when talking about
e.g. about CoreML, int8, fp16, half float, ARKit, AVFoundation, ONNX
etc.
g-mork wrote 23 hours 27 min ago:
How does this compare to Parakeet, which runs wonderfully on CPU?
ac29 wrote 23 hours 31 min ago:
No idea why 'sudo pip install --break-system-packages moonshine-voice'
is the recommended way to install on raspi?
The authors do acknowledge this though and give a slightly too complex
way to do this with uv in an example project (FYI, you dont need to
source anything if you use uv run)
armcat wrote 23 hours 33 min ago:
This is awesome, well done guys, Iâm gonna try it as my ASR component
on the local voice assistant Iâve been building [1] . The tiny
streaming latencies you show look insane
HTML [1]: https://github.com/acatovic/ova
lostmsu wrote 1 day ago:
How does it compare to Microsoft VibeVoice ASR [1] ?
HTML [1]: https://news.ycombinator.com/item?id=46732776
cyanydeez wrote 1 day ago:
No LICENSE no go
altruios wrote 1 day ago:
reading through readme.md
"License
This code, apart from the source in core/third-party, is licensed
under the MIT License, see LICENSE in this repository.
The English-language models are also released under the MIT License.
Models for other languages are released under the Moonshine Community
License, which is a non-commercial license.
The code in core/third-party is licensed according to the terms of
the open source projects it originates from, with details in a
LICENSE file in each subfolder."
bangaladore wrote 1 day ago:
There is a license blurb in the readme.
> This code, apart from the source in core/third-party, is licensed
under the MIT License, see LICENSE in this repository.
> The English-language models are also released under the MIT
License. Models for other languages are released under the Moonshine
Community License, which is a non-commercial license.
> The code in core/third-party is licensed according to the terms of
the open source projects it originates from, with details in a
LICENSE file in each subfolder.
mkl wrote 20 hours 51 min ago:
The LICENSE file that refers to is missing. There's one in the
python folder, but not for the rest of the code.
namibj wrote 15 hours 23 min ago:
IANAL.
Presuming (I haven't checked myself) the git author information
supports this, it should be fine to treat this as licensing the
code it specifies under MIT; based on that license name being (to
my understanding) unambiguous and license application being based
on contract law and contract law basically having at it's very
core the principle of "meeting of the minds" along with wilful
infringement being really really hard to even argue for if the
only thing that's separating it from being 100% clearly licensed
in all proper ways being not copying in an MIT `LICENSE` template
with date and author name pasted into it.
DIR <- back to front page