Processing 174 Hours of NASA Audio With a 0.9B Open-Source Model
vanstriendaniel · x · 2026-07-10
Open MOSS has released MOSS-Transcribe-Diarize, a 0.9B parameter model open-sourced under Apache 2.0 that can perform transcription, speaker diarization, and timestamping in a single run. Using this model, a developer spent less than $10 to process 174 hours of Apollo 11 mission audio from the Internet Archive into searchable data.
Related event: MOSS Releases Open-Source 0.9B Audio Transcription Model Supported by vLLM(5 posts)→
More from Infra
- Two-hour workshop covers open models, benchmark cheating, reward hacking and quantization — danielhanchen · 2026-07-21
- Nativ brings local AI model running to Mac with a desktop app and localhost API — Simon Willison · 2026-07-21
- Octen says agent search now runs at 62ms P50 with only a 6ms P90 gap — aakashgupta · 2026-07-21
- Zhipu acquires a compiler-team spinout to optimize AI inference on domestic chips — zephyr_z9 · 2026-07-21
- Open reproduction of Meta’s REWIRE data pipeline cuts the cost to about $11 — vanstriendaniel · 2026-07-21
- NeurIPS 2026 workshop will focus on on-device intelligence and local execution — YiMaTweets · 2026-07-21