Working on @UnderdogAI’s Woof on my Mac.
This is a AI post classified by AGK Intelligence as Voice AI (a tool drop), kept by the AI Radar because it carries real work, not commentary.
Working on @UnderdogAI’s Woof on my Mac. The whole voice model is 40 MB. It talks about 15x faster than real time on one CPU thread. No GPU! My drama 3 recipe needs 20 GB of GPU memory for about the same accuracy (not expression). Still testing, here is a sample
Posted by netrunner (8.4k followers) 4 h ago · 10 likes · 756 views · view the original post on X. Kept by the AI Radar as Voice AI.
More AI work like this
- OpenAI has added support for outbound phone calls to the GPT-Live API. — @davidarngar
- I honestly didn't think anyone would catch up with ElevenLabs this fast. — @SimonHoiberg
- Underdog Woof TTS could be quite useful: 31.7 MB model with a single voice and with a… — @victormustar
- every night i have a set of 6-10 fun explainers made using raea and ElevenLabs about… — @joelhooks
- You can now send off transcription to select cloud providers if you don't wanna run… — @jullerino
- Woof is 31.7 MB, so small that the whole model runs in your browser in WebGPU — @0xSigil
- Accent matters in Text to Speech. Qwen-Audio-3.1-TTS-Plus leads in US English, while… — @ArtificialAnlys
- Lon recorded a voice note in his car, and his Plaud Agent turned it into a newsletter… — @twistartups
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.2k posts from 5k X accounts over the last 14 days, 10.6k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-07 16:45 UTC. Full method.