feat(smallestai): add endpointing, keyword boosting, format, and sentence timestamps to Smallest STT - #6565
Merged
tinalenguyen merged 5 commits intoJul 31, 2026
Conversation
…ence timestamps to STT Wires up four new Pulse STT streaming options: endpointing (trailing-silence finalization, default true), format (punctuation/capitalization, default true), keywords (word-boost list, default off), and sentence_timestamps (utterances metadata, default off). All four flow through update_options() and trigger a stream reconnect like the existing options.
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
harshitajain165
changed the title
Contributor
Author
|
Hey @tinalenguyen 👋 Just added a few new streaming features to the Smallest AI STT plugin. Would love your eyes on it when you get a chance |
Wires up redact_pii and redact_pci streaming options (default false), matching the format/sentence_timestamps pattern. Redacted entity placeholders from the server surface via SpeechData.metadata["redacted_entities"].
4 tasks
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
endpointing(trailing-silence finalization, defaultTrue),format(punctuation/capitalization, defaultTrue),keywords(word-boost list, default off), andsentence_timestamps(sentence-levelutterancessurfaced viaSpeechData.metadata, default off).redact_piiandredact_pcistreaming options (default off), per the PII and PCI Redaction docs.redact_piimasks names, addresses, and phone numbers;redact_pcimasks credit card numbers, CVVs, ZIP codes, and account numbers. Redacted placeholder tokens (e.g.[FIRSTNAME_1],[CREDITCARDCVV_1]) are surfaced viaSpeechData.metadata["redacted_entities"]. Per the docs, redaction is only reliably supported forlanguage="en"and"hi".STT.update_options()/SpeechStream.update_options()and trigger a reconnect like the existing options (eou_timeout_ms,diarize, etc).utterancesinSpeechData.metadatawheneverword_timestampsis enabled (no separate flag needed there, per the API).Test plan
uv run pytest tests/test_plugin_smallestai_stt.py --unit— 26/26 passingruff check/ruff format --checkon changed files — cleanmypy --strictonstt.py— clean (one pre-existing, unrelated generic-type note on theSTTclass declaration)update_options()reconnect — 17/17 checks passed