Repository navigation
Conversation
Authored with assistance from OpenAI Codex.
Document the tested microphone dashboard, deterministic reference validation, and verified board programming workflow, including observed limits. Authored with assistance from OpenAI Codex.
usamahz
marked this pull request as ready for review
November 18, 2026 14:50
Align the Silero Learning Path metadata with the upstream category validator. Authored with assistance from OpenAI Codex.
usamahz
force-pushed
the
codex/silero-alif-learning-path
branch
from
November 18, 2026 15:43
053356c to
32ccdd1
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add a Learning Path for live Silero VAD on the Alif Ensemble E8 A0, using the on-board microphone, Cortex-M55 HP core, Ethos-U85, and a 480 × 800 dashboard. A separate build validates a deterministic 79-window recording against its host reference before switching to live input. The path includes released dependencies, pinned board support, application sources, and probe-selected programming with fresh MRAM readbacks.
Both firmware configurations cross-built. The physical deterministic run passed all 79 windows with zero threshold-decision mismatches and maximum absolute probability error 4.70 × 10⁻⁷ against rounded UART output; the original tolerances are unchanged. Fresh firmware and signed-package readbacks passed for both modes.
The live acoustic run covered 49.152 seconds at the expected 16 kHz cadence, with speech/quiet response, at most two of eight capture slabs occupied, and no reported errors. The restored live image also passed a separate 20-second reset capture. These sparse observations establish functional response, not an accuracy benchmark. The controlled acoustic interval added no PCM-rail clips to its initial count of 36; the separate final reset increased its count from 38 to 41. The dashboard screenshot is an actual target framebuffer readback with register checks; physical panel updates have not been manually confirmed.
All 15 packaged files match the final source. Nine Markdown pages, 30 Bash blocks, five embedded Python snippets, focused Hugo rendering, and lesson/asset links were checked. Both configurations were built from the packaged sources; the only subsequent archive change was its README.
Authored with assistance from OpenAI Codex.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of the Creative Commons Attribution 4.0 International License.