[pipecat-docs-bot]

Automated documentation update for pipecat PR #6530.

Changes

• api-reference/server/classifiers/hume-audio.mdx (new): HumeAudioClassifier and HumeAudioDecisionsClient; registered in docs.json (Classifiers group) and supported-services.mdx.
• api-reference/server/expressive/overview.mdx: PerceptionReader audio-classifier behavior, timeout now float | None with new defaults, and a PerceptionFrame row in the Frames table.
• api-reference/server/classifiers/overview.mdx: HumeAudioClassifier card; added to the classifiers that open a connection in setup().
• pipecat/learn/classifiers.mdx: noted HumeAudioClassifier as an audio classifier.
• api-reference/server/expressive/pipecat-tone.mdx: PipecatToneReader timeout is now float | None, default None (0.5 seconds).

Gaps identified

None

Automated review

• api-reference/server/expressive/pipecat-tone.mdx: PipecatToneReader timeout documented as float, default 0.5; the PR makes it float | None, default None (0.5 for this text-only reader). Changed to float | None / None and noted 0.5 as the effective default. Caused by this PR's diff; checker is clean for it now.
• reference-findings.txt was empty. Claims in hume-audio.mdx (parameters, defaults, 5s–300s clip limits, 16 kHz, retry statuses, connect/setup behavior, 1000-char context) checked against source: correct.
• Other edited pages (overview, classifiers, supported-services, docs.json): no problems found.

[github-actions]

🔍 Mintlify preview for this branch: https://daily-docs-pr-6530.mintlify.site