Speko
OpenRouter for Voice
In plain words
Speko is an API that unifies speech-to-text, language models, and text-to-speech across multiple providers. It benchmarks 59 voice models across 10 languages and automatically routes each session to the best-performing model for that language, rather than relying on a single top performer. Designed for developers building multilingual voice applications, Speko offers both hosted routing through its infrastructure and an open-source runtime option for frameworks like LiveKit and Pipecat, with a freemium model starting at $0.09 per minute.
written from the facts on this page · September 2026
From the sources
One API for speech-to-text, language models, and text-to-speech, with public benchmarks beside runtime availability.
Every voice model is benchmarked in English. Your users don't speak English. Speko measures 59 speech and language models across 10 languages, then routes each session to the one that actually wins.
Every speech model, benchmarked language by language, wired into one API. A hosted, provider-neutral STT, LLM and TTS data plane at router.speko.dev, with typed contracts and managed routing. Their rank in any other language is unknown — including the model that sits at the top of the English table. No single model is best everywhere, so the right pick changes with the language your users speak. The open customer-side runtime for LiveKit and Pipecat: provider-direct streaming, local BYOK credentials and optional Speko-managed routes. Use the native Gateway integrations in your framework, or call the hosted Router through its public OpenAPI and AsyncAPIfrom speko.ai
Pricing, as stated on its site
Paid from $0.09/mo — Speko offers a freemium model with $100 signup credit for all accounts. Paid plans start at $0.09/minute for Speko infrastructure (covering speech-to-text, LLM, and text-to-speech), Router plan charge
checked September 2026 · prices change
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 9d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com