Pipecat

Add emotion, strength and positivity to each user turn in a Pipecat voice agent with the pipecat-behavioralsignals package (uv add pipecat-behavioralsignals)

Pipecat voice agents can hear how the user speaks, not just what they say. The pipecat-behavioralsignals package streams the user's voice to the Behavioral Signals API and adds what it reads, such as emotion, strength and positivity, to each user turn. The LLM reads it together with the user's words.

Installation

uv add pipecat-behavioralsignals

Or with pip: pip install pipecat-behavioralsignals. Requires Python 3.12 or later and Pipecat 1.16.

Setup

Create an API key in the portal, then set your client ID and API key:

export BEHAVIORALSIGNALS_CID=...
export BEHAVIORALSIGNALS_API_KEY=...

Usage

Put BehavioralSignalsPerceptionService right after your STT service, and turn on expressive conversation:

from pipecat.pipeline.pipeline import Pipeline
from pipecat.pipeline.worker import PipelineParams, PipelineWorker

from pipecat_behavioralsignals import BehavioralSignalsPerceptionService

perception = BehavioralSignalsPerceptionService()

pipeline = Pipeline([
    transport.input(),
    stt,
    perception,
    user_aggregator,
    llm,
    tts,
    transport.output(),
    assistant_aggregator,
])

worker = PipelineWorker(pipeline, params=PipelineParams(enable_expressive_conversation=True))

Links


Did this page help you?