Show HN: Nanosamur.ai – Ollama for Voice Models (STT)

Category: ai-ml

Tags: speech-to-text, orchestration, docker-compose, model-agnostic, privacy

Score: 6.8/10 (Innovation: 6, Technical: 8, Documentation: 7, Utility: 6)

nanosamurai is an open-source, model-agnostic orchestration platform for speech-to-text that supports realtime, refined, and batch processing with multiple models (Whisper, Qwen, Nemotron, Parakeet). It provides a distributed architecture with Kafka, PostgreSQL, and S3-compatible storage, plus a browser UI, Electron app, API, and SDK, aiming to be 'Ollama for voice models' while prioritizing data privacy.

Target audience: backend developers, devops, data engineers, AI/ML engineers

Repository: https://github.com/nanosamurai/nanosamurai · Python · Apache-2.0 · 9 stars

View on Hacker News