From the team
Notes from inside the stack.
Practical writing about private speech AI, reliable transcription systems, and turning sensitive conversations into useful workflows.
-
Model Arena - Qwen3-ASR vs faster-whisper in real time
the WER is not enough: Qwen wins on WER but loses this round regardless
-
Qwen3-ASR joins nanosamur.ai realtime transcription
Run Qwen, Faster-Whisper, or both as independent, labelled realtime tracks from the same audio stream.
-
Why speech AI belongs inside your infrastructure
A practical look at keeping recordings, transcripts, and speech workflows inside infrastructure your organisation controls.
-
From live transcript to trusted final record
Why realtime captions and accurate final transcripts are different products, and how a speech platform can provide both.
-
Designing speaker-aware workflows
How speaker-aware transcripts turn a stream of words into a conversation that people and software can reason about.