Floe

Streaming speech-to-text service over Floe’s transcription WebSocket with per-call spend caps

Overview

FloeSTTService provides streaming speech-to-text through Floe — a unified billing ledger for voice AI where one key powers the LLM, STT, and TTS legs of an agent, metered per call and bounded by pre-call spend caps.

Unlike the LLM and TTS legs (which are OpenAI-compatible base-URL swaps), Floe streaming STT is a dedicated plugin: it opens a WebSocket to Floe, streams raw PCM audio up, and receives JSON transcript messages back. It extends Pipecat’s WebsocketSTTService (the same base used by the Deepgram and Gladia services), so connect/receive/reconnect scaffolding, audio buffering, and metrics are inherited. Non-final results are emitted as InterimTranscriptionFrame and final results as TranscriptionFrame.

Source Repository

Source code, examples, and issues for the Floe integration

PyPI Package

The pipecat-floe package on PyPI

Floe

Documentation and supported models for Floe

API Keys

Create and manage your Floe agent keys

Installation

This is a community-maintained package distributed separately from pipecat-ai:

uv add pipecat-floe

Prerequisites

Floe Account Setup

Before using the Floe STT service, you need:

1. Floe Account: Sign up at the Floe dashboard
2. API Key: Create an agent key from your dashboard. New keys come with welcome credit that covers the first calls.

Required Environment Variables

• FLOE_API_KEY: Your Floe agent key for authentication

Configuration

api_key · str · default: "None"

Floe agent key. Falls back to the FLOE_API_KEY environment variable if not provided. Raises ValueError if neither is set.

model · str · default: "deepgram/nova-3"

Fully qualified provider/model STT identifier (for example "deepgram/nova-3").

encoding · str · default: "linear16"

Audio encoding of the PCM frames streamed to Floe. One of "linear16", "mulaw", "alaw". Pipecat transports deliver linear16 PCM by default.

sample_rate · int · default: "16000"

Audio sample rate in Hz (8000–48000).

language · str · default: "en"

BCP-47 language hint.

base_url · str

Floe streaming-STT WebSocket URL.

kwargs

Additional keyword arguments passed through to the underlying WebsocketSTTService.

Usage

from pipecat_floe import FloeLLMService, FloeSTTService, FloeTTSService
from pipecat.pipeline.pipeline import Pipeline

# One Floe key powers all three legs, on a single budget.
stt = FloeSTTService(model="deepgram/nova-3")       # streaming STT (WebSocket)
llm = FloeLLMService(model="openai/gpt-4o-mini")
tts = FloeTTSService(model="openai/tts-1", voice="alloy")

# Build pipeline
pipeline = Pipeline([stt, llm, tts])

See the source repository for a complete runnable example wiring a WebSocket transport → STT → LLM → TTS.

Compatibility

Built and verified against Pipecat v1.7.0+. Check the source repository for the latest tested version and changelog.