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

Launch HN: Speko (YC S26) – OpenRouter for Voice AI

speko.ai

Post reply on HN
Print view View original

31–40 of 73 posts

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#31

Post by jakswa »

Tepix wrote: post #23

Ever since OpenAI launched their improved voice mode, I've been looking for a capable LLM with builtin voice-in and -out. Llama 4 was supposed to be it but turned out to be a dud. I haven't followed the topic closely lately, did I miss anything? Are there capable (!) open weights omni models that allow low latency voice chat? If so, what software do you use them with? Can you use a PWA on your phone? WebRTC? WebTrans…

Gemma 4 (both E4B + 12B) performed really well as ears+brains. I mostly comment because I too am always scouting for a nice local all-in-one model.

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#32

Post by abdik »

IgorBlink wrote: post #27

any progress with on-device models??

there is a good progress on on-devise models, but not ready for production yet to fit in devices. But as soon as there is are some good results, we are going to benchmark them and put in https://benchmarks.speko.ai/

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#33

Post by cootsnuck »

jeffrwells wrote: post #29
maho wrote: post #26

Which model best allows me to transcribe speech that uses a lot of domain-specific terms? For example, when I say "Claude Code", it often gets transcribed as "Cloud Code", and I have to go back and edit or do a second pass with a traditional LLM (which can introduce additional errors).

I’ve had a lot of success in the past with fine tuning STT using synthetic data. I was doing it for Veterinary (ambient recording -> SOAP notes) which has tons of complex domain-specific language AND it is critically important to get right. “CPR” transcribing as “see pee are” just doesn’t cut it in that industry.

Which open source STT models have you had success with for fine tuning?

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#35

Post by sparkling »

alabhyajindal wrote: post #20
sparkling wrote: post #19

Earlier quoted context omitted.

Just canceled my WisprFlow subscription a few days ago to switch to a open source, free, local alternative. In my case handy.computer with the Cohere model.

You mean handy.computer?

Yup, my bad, fixed.

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#36

Post by mlhpdx »

Tepix wrote: post #23

Ever since OpenAI launched their improved voice mode, I've been looking for a capable LLM with builtin voice-in and -out. Llama 4 was supposed to be it but turned out to be a dud. I haven't followed the topic closely lately, did I miss anything? Are there capable (!) open weights omni models that allow low latency voice chat? If so, what software do you use them with? Can you use a PWA on your phone? WebRTC? WebTrans…

FWIW, I have seen a couple phone apps using react native and sending directly to backends using a socket. Seems to avoid some overhead but I don’t have first hand experience with the client side.

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#37

Post by cootsnuck »

narrationbox wrote: post #7

> Typical production voice agent is an ensemble of three models: STT, an LLM, and TTS. To use a claudism , I would like to push back on this. The industry is very much moving towards one-model-does-all end to end trained similar to LLMs and VLMs. Mostly for latency reasons and partially because the results for the end to end trained models are just so much better than those using three pieces architectures. I think m…

> The industry is very much moving towards one-model-does-all end to end trained

I've worked with hundreds of enterprises on voice AI and voice agent solutions. In my experience, this isn't true. Or rather I should say, the people actually paying for voice agents (i.e enterprises) are not moving towards STS solutions in a meaningful way. The composability, observability, and reliability profile of STS systems is not amenable to enterprise criteria. Not to mention costs.

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#38

Post by k9294 »

maho wrote: post #26

Which model best allows me to transcribe speech that uses a lot of domain-specific terms? For example, when I say "Claude Code", it often gets transcribed as "Cloud Code", and I have to go back and edit or do a second pass with a traditional LLM (which can introduce additional errors).

Gemini 3 flash series is quite good, also elevenlabs scribe v2

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#39

Post by abdik »

spmartin823 wrote: post #11

Does this include a turn taking API? It'd be great to have one API that could do "Conversation in a box". One of the biggest annoyances is daisy chaining many models together for turn taking, dumb models for immediate responses, with smarter models returning and taking over after.

Yes, on the hosted side (agents platform): full sessions come with VAD and turn-taking handled - we set them up and tune them for your use case, so that is the closest thing to conversation in a box. If you run your own orchestration, the gateway is just the routing layer and turn-taking stays in your framework - in our own stack we run Pipecat's Smart Turn in-process and tune the commit threshold on real calls. We also share our benchmarks here: https://benchmarks.speko.ai/turntaking

On fast dumb models answering while a smarter one takes over: we are experimenting with exactly that split - a small fast model holds the conversation while a larger one works behind it. Today it runs as two pinned routes, not one packaged API. Most turns in a phone call do not need a frontier model, and the fastest models on our LLM board are all small, so this is where routing earns its keep. We publish benchmarks on LLMs here: https://benchmarks.speko.ai/llm

Re: Launch HN: Speko (YC S26) – OpenRouter for Voice AI

#40

Post by abdik »

webo wrote: post #17

The benchmarks page seems interesting and something I can use to help make an informed decision. Can you talk about how you're measuring some of these? I imagine it needs to involve some human input. https://benchmarks.speko.ai/turntaking

Turn-taking specifically does not need a listening panel, but it is measured mechanically. 200+ real human clips, and we score end-vs-wait decisions: did the model decide the caller finished speaking, or just paused mid-thought. The best detector gets 94.0% of those right; a plain VAD silence timer gets 46.9%. Results are published here: https://benchmarks.speko.ai/turntaking

You are right about human input for naturalness, that one we did not automate away with yet. We run blind A/B listening rounds with native speakers.

Post reply on HN