feat(transcription): Parakeet-v3 model download (Track-B step 3)#11
Open
farce1 wants to merge 1 commit into
Open
feat(transcription): Parakeet-v3 model download (Track-B step 3)#11farce1 wants to merge 1 commit into
farce1 wants to merge 1 commit into
Conversation
…all) Track-B step 3. Adds model::parakeet_model_dir + check_parakeet_assets_ready, PARAKEET_V3_ARCHIVE (k2-fsa asr-models tarball), download_parakeet_model (mirrors the whisper-turbo flow: range-resume download + SHA256 verify + tar.bz2 extract + readiness check), the download_parakeet_model command, and its registration. SHA256 is a REPLACE_WITH_ placeholder with a fail-loud test, per the existing maintainer-fill convention.
This was referenced Jun 14, 2026
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What — Track-B step 3 (stacked on #10)
Backend download for the Parakeet-v3 model so the engine from #10 has files to load.
model.rs:parakeet_model_dir+check_parakeet_assets_ready(encoder/decoder/joiner.int8.onnx + tokens.txt) — with a unit test.download.rs:PARAKEET_V3_ARCHIVE(k2-fsaasr-modelstarball, verified URL…/sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8.tar.bz2) +download_parakeet_modelmirroring the Whisper-turbo flow (range-resume download → SHA256 verify →extract_tar_bz2→ readiness) + a fail-loud SHA test.commands.rs+lib.rs:download_parakeet_modelcommand, registered.The extracted dir matches what
ParakeetEngine::load(#10) expects.Verification
clippy --all-targets --all-features -- -D warnings✅ (full chain compiles),model::tests✅.PARAKEET_V3_ARCHIVE.sha256is aREPLACE_WITH_placeholder with a fail-loud test — same maintainer-fill convention as the Whisper/diarization archives (now 3 such gated SHAs). Runtime download untestable in CI (network + 671 MB model).Next (Track-B)
Engine-selection setting (default Parakeet-v3, Whisper fallback) — plumb
asr_engine+ per-engine model dir throughStartWorkerArgs/session.rs/commands.rs + a first-run/download trigger; then picker UI; GPU EPs;retranscribe_meeting.Notes
Stacked on #10 (base
feat/parakeet-engine); retarget chain to main on merge. CI red on the usual maintainer-only blockers, not this PR.