Voice gender classifier for European voice AI (1MB, ONNX, 4ms)
Hi, I'm Kamil and I'm a founder of Applied AI agency in Warsaw, Poland. We've trained a small <1MB voice classifier model that runs on CPU in 4ms. Can be run next to silero VAD in voice AI deployments. What we noticed in production deployments of voice assistants in Contact Centers in EU is that human consultants pick up immediately how to inflect verbs and ajdectives after one utterance from the caller. But voice AI agents don't know it until 1-2 minutes into the call when they are either corrected or the caller uses explicitly words with male/female form a couple of times. Our model…
What it does
In the maker’s words, at launch
Hi, I'm Kamil and I'm a founder of Applied AI agency in Warsaw, Poland. We've trained a small <1MB voice classifier model that runs on CPU in 4ms. Can be run next to silero VAD in voice AI deployments. What we noticed in production deployments of voice assistants in Contact Centers in EU is that human consultants pick up immediately how to inflect verbs and ajdectives after one utterance from the caller. But voice AI agents don't know it until 1-2 minutes into the call when they are either corrected or the caller uses explicitly words with male/female form a couple of times. Our model solves this just from the first utterance of caller speech. The voice AI pipeline can inject the classification as context to the system prompt. We observed a significant impact of this on the adoption of voice AI in practice. Model + paper: https://huggingface.co/syntropicsignal-ai/gender-voice-class...
Does the same job
all alternatives →
Lucas | Call your AI language teacher2024 · ▲204Learn English, Spanish, German, French & more by talking
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 · 17d 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 · 27d ago · cactuscompute.com
Launched alongside, May 2026
the whole month →
Parallel agents, diff reviewer, and multi-model comparisons
Dev tools · May 2026 · kilo.ai
- NW
Hey HN, Henry here from Cactus. We open-sourced Needle, a 26M parameter function-calling (tool use) model. It runs at 6000 tok/s prefill and 1200 tok/s decode on consumer devices. We were always frustrated by the little effort made towards building agentic models that run on budget phones, so we conducted investigations that led to an observation: agentic experiences are built upon tool calling, and massive models are overkill for it. Tool calling is fundamentally retrieval-and-assembly (match query to tool name, extract argument values, emit JSON), not reasoning. Cross-attention…
Life & fun · May 2026 · github.com
- FM
Dev tools · May 2026 · github.com