sttlab-tech 936f1b6402 serialize ASR across tracks onto one shared worker, cut 1.7B peak memory ~13GB->~7GB ...

qwen_asr has no quantized-weights option, so the only lever available to
reduce the 1.7B model's memory footprint is avoiding concurrent instances.
Previously each track (system, mic) ran its own transcribeTrack goroutine
with independent qwen_asr subprocesses, so simultaneous speech on both
tracks meant two ~6.9GiB model instances alive at once - over half the
24GB target machine's memory. Merge both tracks onto a single worker
(asrSegs channel, dispatch() routes both tracks onto it) so invocations
are strictly serialized; per-track order is preserved since VAD's output
is already chronological. Verified via pgrep -x qwen_asr polling during a
real dual-track capture: never more than 1 concurrent process.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

2026-08-10 11:30:23 +02:00

..

transcript-tail

initial commit: live meeting transcription pipeline

2026-08-09 13:18:57 +02:00

transcriptor-ai

serialize ASR across tracks onto one shared worker, cut 1.7B peak memory ~13GB->~7GB

2026-08-10 11:30:23 +02:00