Krisp VIVA
Krisp's voice isolation filter, VAD analyzer and turn analyzer. Deferred in pipedog: not ported yet.
Deferred
Krisp VIVA is deferred: it is possible, but it isn't ready to port.
Why. The Krisp VIVA SDK is proprietary. It isn't on public npm, and it needs a license and the model files from Krisp.
What would unblock it. Bring your own SDK: core already has the KrispAudioSdk seam (audio/krisp_instance), and setKrispAudioSdk() takes the SDK you obtained from Krisp. An optional package that wires it up is planned.
Sources: Krisp VIVA in Pipecat.
What it is
Krisp's voice isolation filter, VAD analyzer and turn analyzer.
Upstream, it's these classes: KrispVivaFilter, KrispVivaVadAnalyzer, KrispVivaTurn, in
audio/filters/krisp_viva_filter.py and audio/vad/krisp_viva_vad.py and audio/turn/krisp_viva_turn.py.
Its pipedog package will be @pipedog/audio-krisp-viva-filter.
Get it ported
- Search the issues for a request, and upvote it with a reaction.
- If there's none, open a port request.
- Or port it yourself: the contributing guide explains how modules are ported, test by test. The porting status lists what's done.
Until then, see what is ported in Audio filters & mixers.
pipedog is an unofficial community TypeScript port of Pipecat. It is not affiliated with or endorsed by Daily or the Pipecat project.
Portions adapted from the Pipecat documentation, Copyright (c) 2024–2025 Daily, BSD-2-Clause (see apps/docs/LICENSE-pipecat-docs).