PhD | AI & Voice | Startups https://xitter.cf/t.co/PxzriWj2jt
Paris, Île-de-France, FRA
Joined April 2013
- Tweets13.3K
- Following688
- Followers3.7K
- Likes26.5K
Josh Meyer retweeted
VAmoS Pro puts two to four requests in one call. Its paper describes agents missing a confirmed action when the caller moves to the next request. A useful test for voice systems: does conversational progress preserve unfinished commitments, or quietly erase them?
New on the VAmoS pareto frontier: @smallest_AI 🚀
We ran Smallest's full voice stack (STT + LLM + TTS) on 300 realistic, difficult calls
TLDR;
-> 26.8% task completion at $0.048 a call
-> on cost/completion frontier, pushing @huggingface off
links in comments 🔗
Josh Meyer retweeted
Open roles:
→ Head of Marketing: join.com/companies/ai-cousti…
→ Junior Growth Marketer: join.com/companies/ai-cousti…
→ GTM Growth Manager: join.com/companies/ai-cousti…
→ Forward Deployed Engineer: join.com/companies/ai-cousti…
Apply through the links, or tag someone you think would fit.
Josh Meyer retweeted
.@smallest_AI is the best full stack platform to build your voice agents
New on the VAmoS pareto frontier: @smallest_AI 🚀
We ran Smallest's full voice stack (STT + LLM + TTS) on 300 realistic, difficult calls
TLDR;
-> 26.8% task completion at $0.048 a call
-> on cost/completion frontier, pushing @huggingface off
links in comments 🔗
Josh Meyer retweeted
try our stack, you'll enjoy our llm!
New on the VAmoS pareto frontier: @smallest_AI 🚀
We ran Smallest's full voice stack (STT + LLM + TTS) on 300 realistic, difficult calls
TLDR;
-> 26.8% task completion at $0.048 a call
-> on cost/completion frontier, pushing @huggingface off
links in comments 🔗
Josh Meyer retweeted
The most efficient stack to build Voice Agents - Smallest AI
New on the VAmoS pareto frontier: @smallest_AI 🚀
We ran Smallest's full voice stack (STT + LLM + TTS) on 300 realistic, difficult calls
TLDR;
-> 26.8% task completion at $0.048 a call
-> on cost/completion frontier, pushing @huggingface off
links in comments 🔗
New on the VAmoS pareto frontier: @smallest_AI 🚀
We ran Smallest's full voice stack (STT + LLM + TTS) on 300 realistic, difficult calls
TLDR;
-> 26.8% task completion at $0.048 a call
-> on cost/completion frontier, pushing @huggingface off
links in comments 🔗
🔗
leaderboard: veris.ai/leaderboard/vamos-p…
agent source code: github.com/veris-ai/rory-age…
VAmoS paper: arxiv.org/abs/2607.27453
Josh Meyer retweeted
Fish Speech Drama 3 deep dive: 35.9B open-weight MoE TTS, ~3B active. 7.8x the size of Fish S2, yet faster: 271 vs 177 frames/s on one H200.
Why: 6.0 GB of weight reads per frame vs 15.6 GB. 13 serial layer passes after the backbone vs 40.
erogol.substack.com/p/fish-a…
Josh Meyer retweeted
heading to @money2020 today. DM me if you are in Vegas and want to chat.
TIL; safety and liveliness in software engineering: en.wikipedia.org/wiki/Safety…
ngl, openai dots is pretty awesome. i see a not-to-distant future where i will never open my email again
Josh Meyer retweeted
ELARSpeech: A 30K-Hour Dataset Across 675 Underrepresented Languages for Long-Tail Multilingual Speech Technologies
ELARSpeechは675の少数言語にわたる30,202時間の音声データセットで、ASRとTTSの拡張を支援する。
arxiv.org/abs/2610.19388
This is a super cool bench 🔥
Voice agents can talk with you one-on-one.
But can they join a group conversation? Not yet!
(1/4)
In MP-Bench (#EMNLP2026), 12 voice agents are no better than a coin flip at knowing when to speak, unless they're called by name. 🧵