Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Voxtral Transcribe 2

mistral.ai

Post reply on HN
Print view View original

171–180 of 255 posts

Re: Voxtral Transcribe 2

#171

Post by iagooar »

DaedalusII wrote: post #168
iagooar wrote: post #81

In English it is pretty good. But talk to it in Polish, and suddenly it thinks you speak Russian? Ukranian? Belarus? I would understand if an American company launched this, but for a company being so proud about their European roots, I think it should have better support for major European languages. I tried English + Polish: > All right, I'm not really sure if transcribing this makes a lot of sense. Maybe not. A ць…

polish logically should be rendered in cyrillic as the cyrillic orthography more closely matches the sounds and consonant structure of slavic languages like polish and russian, although this has never been done for church reasons . maybe this is confusing ai

Polish has been written with Latin alphabet since the 13th century. And before it simply wasn't written.

Polish works with the Latin alphabet just fine.

"Do kraju tego, gdzie kruszynę chleba podnoszą z ziemi przez uszanowanie dla darów Nieba.... Tęskno mi, Panie..."

"Mimozami jesień się zaczyna, złotawa, krucha i miła. To ty, to ty jesteś ta dziewczyna, która do mnie na ulicę wychodziła."

Re: Voxtral Transcribe 2

#172

Post by janalsncm »

popalchemist wrote: post #78
janalsncm wrote: post #53

I noticed that this model is multilingual and understands 14 languages. For many use cases, we probably only need a single language, and the extra 13 are simply adding extra latency. I believe there will be a trend in the coming years of trimming the fat off of these jack of all trades models. https://aclanthology.org/2025.findings-acl.87/

It doesn't make sense to have a language-restricted transcription model because of code switching. People aren't machines, we don't stick to our native languages without failure. Even monolingual people move in and out of their native language when using "borrowed" words/phrases. A single-language model will often fail to deal with that.

Everything is a tradeoff, and different use cases require different tradeoffs:

Option A: this model

Option B: faster model, only 1 language

Option C: same size model, only 1 language but higher quality

My point is that option A isn’t always best.

And on the borrowed words bit, there’s no rule that we cannot add borrowed words into the vocab. But you don’t need the whole language. I know what deja voux means but I don’t speak French.

Re: Voxtral Transcribe 2

#173

Post by janalsncm »

m463 wrote: post #137
janalsncm wrote: post #53

I noticed that this model is multilingual and understands 14 languages. For many use cases, we probably only need a single language, and the extra 13 are simply adding extra latency. I believe there will be a trend in the coming years of trimming the fat off of these jack of all trades models. https://aclanthology.org/2025.findings-acl.87/

I don't know. What about words inherited from other languages? I think a cross-language model could improve lots of things. For example, "here it is, voila!" "turn left on el camino real"

Most English speakers likely would understand those and don’t speak French or Spanish. So it’s not necessary to tack on extra languages even if there are loan words.

In general there is a concept called the “curse of multilinguality”

https://arxiv.org/pdf/1911.02116

Re: Voxtral Transcribe 2

#175

Post by gerad »

timhh wrote: post #153
londons_explore wrote: post #146

Earlier quoted context omitted.

200 million people speak Swahili. 39 million people speak Polish, and most of those also speak English or another more common language.

You could say the same about Dutch to be fair. 90-95% speak English - I bet that's way higher than in Poland.

As an American, my perspective is that Dutch people speak better English than a large percentage of English people and Americans.

Re: Voxtral Transcribe 2

#176

Post by GolDDranks »

simonw wrote: post #20

This demo is really impressive: https://huggingface.co/spaces/mistralai/Voxtral-Mini-Realtim... Don't be confused if it says "no microphone", the moment you click the record button it will request browser permission and then start working. I spoke fast and dropped in some jargon and it got it all right - I said this and it transcribed it exactly right, WebAssembly spelling included: > Can you tell me about RSS and At…

I can't get that demo to work. Tried with both Firefox and Chrome.

Re: Voxtral Transcribe 2

#177

Post by harry8 »

What hardware resources are required for what quality/latency? Multiple high end nvidia or can you run it on your phone on an esp32 offline? Or...

Seems like fundamental info for any model announcement. Did I just miss it? Does everyone just know except me?

Re: Voxtral Transcribe 2

#178

Post by janalsncm »

keeganpoppen wrote: post #91
janalsncm wrote: post #53

I noticed that this model is multilingual and understands 14 languages. For many use cases, we probably only need a single language, and the extra 13 are simply adding extra latency. I believe there will be a trend in the coming years of trimming the fat off of these jack of all trades models. https://aclanthology.org/2025.findings-acl.87/

uhhh i cast doubt on multi-language support as affecting latency. model size, maybe, but what is the mechanism for making latency worse? i think of model latency as O(log(model size))… but i am open to being wrong / that being a not-good mental model / educated guess.

Well for example the last step is to softmax over all output logits, which is the same as your vocab size. You need the sum of the exponentiated values of each logit to calculate the denominator which is O(N).

Bigger impact is before that you need to project the hidden state matrix to the vocab list. Something like 4096x250000. Bigger vocab=more FLOPs.

If you’re on a GPU things are parallelized so maybe it’s not quite linear if everything fits nicely. But on a cpu you’re going to struggle more.

This is why the juiciest target when shrinking models is the token embedding table. For example AlBERT factorized the whole embedding table to two low rank matrices.

Re: Voxtral Transcribe 2

#180

Post by black_puppydog »

janalsncm wrote: post #53

I noticed that this model is multilingual and understands 14 languages. For many use cases, we probably only need a single language, and the extra 13 are simply adding extra latency. I believe there will be a trend in the coming years of trimming the fat off of these jack of all trades models. https://aclanthology.org/2025.findings-acl.87/

honestly the inability to correctly transcribe the 4 language mix i use in my everyday life is one of the major blockers for adopting ASR tech in my own tooling. this coming from someone who literally works in that field.

turns out, outside the US, many people speak more than one language. :)

edit: I should say was a major blocker, because the last iterations of open-weight models actually work better and better. it's often the UX that's not thought for these usecases.

Post reply on HN