[HN Gopher] Cohere Transcribe: Speech Recognition
       ___________________________________________________________________
        
       Cohere Transcribe: Speech Recognition
        
       Author : gmays
       Score  : 145 points
       Date   : 2026-03-31 16:27 UTC (6 hours ago)
        
 (HTM) web link (cohere.com)
 (TXT) w3m dump (cohere.com)
        
       | geooff_ wrote:
       | I can't say enough nice things about Cohere's services. I
       | migrated over to their embedding model a few months ago for clip-
       | style embeddings and it's been fantastic.
       | 
       | It has the most crisp, steady P50 of any external service I've
       | used in a long time.
        
         | bluegatty wrote:
         | can u comment on overall quality? their models tend to be a bit
         | smaller and less performant overall.
        
           | geooff_ wrote:
           | My baseline was Jina, A Chinese model provider. I had major
           | issues with their reliability. I have no comparison to
           | provide in terms of offline metrics as I had to do an
           | emergency migration because their inference service has
           | extended downtimes.
           | 
           | My experience with Cohere and interacting with their sales
           | engineers has been boring, I say that is the most flattering
           | way possible. Embeddings are a core service at this point
           | like VMs and DBs. They just need to work and work well and
           | thats what they're selling.
        
       | simonw wrote:
       | It's great that this is Apache 2.0 licensed - several of Cohere's
       | other models are licensed free for non-commercial use only.
        
       | dinakernel wrote:
       | My worry is that ASR will end up like OCR. If the multi modal
       | large AI system is good enough (latency wise), the advantage of
       | domain understanding eats the other technlogies alive.
       | 
       | In OCR, even when the characters are poorly scanned, the deep
       | domain understanding these large multi modal AIs have allows it
       | to understand what the document actually meant - this is going to
       | be order id because in the million invoices I have seen before
       | order id is normally below order date - etc. The same issue is
       | going to be there in ASR also is my worry.
        
         | progbits wrote:
         | This is both good and bad. Good ASR can often understand low
         | quality / garbled speech that I could not figure out, but it
         | also "over corrects" sometimes and replaces correct but low
         | prior words with incorrect but much more common ones.
         | 
         | With OCR the risk is you get another xerox[1] incident where
         | all your data looks plausible but is incorrect. Hope you kept
         | the originals!
         | 
         | (This is why for my personal doc scans, I use OCR only for full
         | text search, but retain the original raw scans forever)
         | 
         | [1] https://www.dkriesel.com/en/blog/2013/0802_xerox-
         | workcentres...
        
         | nkzd wrote:
         | Why are you 'worried' about it? Shouldn't we strive for better
         | technology even if it means some will 'lose'?
        
           | yorwba wrote:
           | "Better" isn't just about increasing benchmark numbers.
           | Often, it's more important that a system fails safely than
           | how often it fails. Automatic speech recognition that guesses
           | when the input is unclear will occasionally be right and
           | therefore have a lower word error rate, but if it's important
           | that the output be correct, it might be better to insert
           | "[unintelligible]" and have a human double-check.
        
           | IshKebab wrote:
           | It's better in terms of WER. It's not better in terms of not
           | making shit up that sounds plausible.
           | 
           | Probably the answer is simply to tweak the metric so it's a
           | bit more smart than WER - allow "unclear" output which is
           | penalised less than actually incorrect answers. I'd be
           | surprised if nobody has done that.
        
           | ks2048 wrote:
           | Ideally, you'd be able to specify exactly what you want - do
           | you want to write-out filled pauses ("aaah", "umm")? Do you
           | want to get a transcription of the the disfluencies - re-
           | starts, etc. or just get out a cleaned up version?
        
         | corlinp wrote:
         | This is exactly the case today. Multimodal LLMs like
         | gpt-4o-transcribe are way better than traditional ASR, not only
         | because of deeper understanding but because of the ability to
         | actually prompt it with your company's specific terminology,
         | org chart, etc.
         | 
         | For example, if the prompt includes that Caitlin is an
         | accountant and Kaitlyn is an engineer, if you transcribe "Tell
         | Kaitlyn to review my PR" it will know who you're referring to.
         | That's something WER doesn't really capture.
         | 
         | BTW, I built an open-source Mac tool for using
         | gpt-4o-transcribe with an OpenAI API key and custom prompts:
         | https://github.com/corlinp/voibe
        
           | Bolwin wrote:
           | Many ASR models already support prompts/adding your own
           | terminology. This one doesn't, but full LLMs especially such
           | expensive ones aren't needed for that.
        
         | regularfry wrote:
         | For quite a long time there will be a greater advantage to
         | local processing for STT than for TTT chat, or even OCR. Being
         | able to do STT on the device that owns the microphone means
         | that the bandwidth off that device can be dramatically reduced,
         | if it's even necessary for the task at hand.
        
       | topazas wrote:
       | How hard could it be to train other European language(-s)?
        
         | gunalx wrote:
         | If you have to ask you dont really need the answer.
         | 
         | Seems to not be to difficult in finding or creating training
         | code. So a pretty decent amount of high quality training data
         | should be many hours. And a few hours in high end data enter
         | GPU compute, and many iterations to get it right.
        
         | harvey9 wrote:
         | It includes several European languages.
        
           | stronglikedan wrote:
           | hence "other" lol
        
       | teach wrote:
       | Dumb question, but if this is "open source" is there source code
       | somewhere? Or does that term mean something different in the
       | world of models that must be trained to be useful?
        
         | stronglikedan wrote:
         | I presume it means the model itself.
        
         | Doman wrote:
         | Files can be downloaded here:
         | https://huggingface.co/CohereLabs/cohere-transcribe-03-2026/...
         | 
         | And someone has already converted it to onnx format:
         | https://huggingface.co/eschmidbauer/cohere-transcribe-03-202...
         | - so it can be run on CPU instead of GPU.
        
         | gunalx wrote:
         | Most use definition is just awailable weigths.
         | 
         | This kids make sense because "compiling" (training) the model
         | cost inhibitly much, and we can still benefit from the
         | artifacts.
        
       | gruez wrote:
       | > Limitations
       | 
       | >Timestamps/Speaker diarization. The model does not feature
       | either of these.
       | 
       | What a shame. Is whisperx still the best choice if you want
       | timestamps/diarization?
        
         | akreal wrote:
         | WhisperX is not a model but a software package built around
         | Whisper and some other models, including diarization and
         | alignment ones. Something similar will be built around the
         | Cohere Transcribe model, maybe even just an integration to
         | WhisperX itself.
        
         | GaggiX wrote:
         | There is also: https://github.com/linto-ai/whisper-timestamped
         | 
         | It doesn't use an extra model (so it supports every language
         | that works with Whisper out of the box and use less memory), it
         | works by applying Dynamic Time Warping to cross-attention
         | weights.
        
           | oezi wrote:
           | Just a warning that plain WhisperX is more accurate and
           | Whisper-timestamped has many weird quirks.
        
         | bartman wrote:
         | Even in the commercial space, there's a lack of production
         | grade ASR APIs that support diarization and word level
         | timestamps.
         | 
         | My experiences with Google's Chirp have been horrendous, with
         | it sometimes skipping sections of speech entirely,
         | hallucinating speech where the audio contains noise, and
         | unreliable word level timestamps. And this all is even with
         | using their new audio prefiltering feature.
         | 
         | AWS works slightly better, but also has trouble with keeping
         | word level timestamps in sync.
         | 
         | Whisper is nice but hallucinates regularly.
         | 
         | OpenAI's new transcription models are delivering accurate
         | output but do not support word level timestamps...
         | 
         | A lot of this could be worked around by sending the resulting
         | transcripts through a few layers of post processing, but... I
         | just want to pay for an API that is reliable and saves me from
         | doing all that work.
        
           | stavros wrote:
           | Isn't Elevenlabs the best in this?
        
             | bartman wrote:
             | I've not tested their speech-to-text yet, but based on the
             | docs it looks promising. Thanks for the suggestion!
        
               | stavros wrote:
               | It's fantastic, and their diarization is spot on as well.
        
         | lifesaverluke wrote:
         | For podcasts there is this
         | https://news.ycombinator.com/item?id=47584376
        
         | atoav wrote:
         | I would try Qwen-ASR: https://qwen.ai/blog?id=qwen3asr
         | 
         | See the very bottom of the page for a transcription with
         | timestamps.
        
       | Void_ wrote:
       | Just today I shipped support for this in Whisper Memos:
       | https://whispermemos.com/changelog/2026-04-cohere-transcribe
       | 
       | Accurate and fast model, very happy with it so far!
        
       | ramon156 wrote:
       | I had to set-up fireflies for our company recently. Cool tool,
       | but I'm sending dozens of internal meetings to an american
       | company. Our ISO inspector wouldn't be pleased to know.
       | 
       | This is a good option. Will check it out.
        
         | Oras wrote:
         | There are many open source STT models that can run locally on
         | Mac with good performance, such as whisper and Parakeet
        
       | stavros wrote:
       | To clarify, this is SOTA in its size category, right? It's not
       | better than Parakeet, for example?
        
         | caminanteblanco wrote:
         | Well, to clarify, it is both larger than parakeet in parameter
         | count (parakeet is available in 0.6B and 1.1B), since it's 2B
         | params, and also performs better than it on the benchmarks that
         | hugging face publishes on the openASR leaderboard
        
           | stavros wrote:
           | Ahh thanks, I confused my parameter count, thanks. I guess
           | Parakeet is 0.6B, I was somehow thinking 6B.
        
         | jwineinger wrote:
         | Looking at the ASR leaderboard
         | (https://huggingface.co/spaces/hf-audio/open_asr_leaderboard),
         | Parakeet (.6B) is still near the top on speed, but about 10th
         | on WER.
        
           | stavros wrote:
           | Thanks, I don't know how much to trust benchmarks so I
           | figured I'd ask.
        
       | kalmuraee wrote:
       | Multimodels are way better
        
         | Fidelix wrote:
         | Can you clarify? I tested a few and they are rubbish and don't
         | have the same features.
        
       | _medihack_ wrote:
       | Unfortunately, this model does not seem to support a custom
       | vocabulary, word boosting or an additional prompt.
        
       | kieloo wrote:
       | The problem with many STT models is that they seem to mostly be
       | trained on perfectly-accented speech and struggle a lot with
       | foreign accents so I'm curious to try this one as a Frenchman
       | with a rather French English accent.
       | 
       | So far, the best I have found while testing models for my
       | language learning app (Copycat Cafe) is Soniox. All others
       | performed badly for non native accents. The worst were whisper-
       | based models because they hallucinate when they misunderstand and
       | tend to come up with random phrases that have nothing to do with
       | the topic.
        
       | BreezyBadger wrote:
       | Awesome. Going to see if I can port https://scrivvy.ai to this.
       | based in Canada
        
       | bkitano19 wrote:
       | notable omission of deepgram models in comparisons?
        
       | mnbbrown wrote:
       | Ran it over our internal dataset of ~250 recordings of people
       | saying british postcodes (all kinds of accents, etc) - it's
       | competitive for sure!
       | 
       | Soniox (stt-async-v4): 176/248 (71.0%) ElevenLabs (scribe_v2):
       | 170/248 (68.5%) AssemblyAI (universal-3-pro): 166/248 (66.9%)
       | Deepgram (nova-3): 158/248 (63.7%) AssemblyAI (universal-2):
       | 148/248 (59.7%) Cohere (transcribe-03-2026): 148/248 (59.7%)
       | Speechmatics (enhanced): 134/248 (54.0%)
       | 
       | P.s. how do I get this to render correctly on here?
        
         | Bolwin wrote:
         | Try two newlines between each one
        
       ___________________________________________________________________
       (page generated 2026-03-31 23:00 UTC)