Pipecat
Saidwire speech-to-text as a Pipecat STT service for voice agents.
pip install pipecat-saidwireIntegrations
Works with the tools you already use.
Plug-ins for Pipecat, LiveKit, the Vercel AI SDK and MCP, plus setup guides for tools that speak the OpenAI transcription API or Deepgram's streaming API. Most tools work after you change the base URL and the key.
LiveKit
A LiveKit Agents STT plug-in for real-time voice agents.
pip install livekit-plugins-saidwireVapi
Voice-agent platform. A Saidwire adapter is coming.
Dify
Speech-to-text in Dify through the OpenAI-compatible endpoint.
Flowise
Transcribe audio in Flowise flows through the OpenAI-compatible endpoint.
Vercel AI SDK
A Saidwire provider for the AI SDK's transcription function, or the OpenAI provider with our base URL.
npm install @saidwire/ai-sdk-providern8n
A Saidwire community node for n8n workflows. In n8n: Settings → Community Nodes → install n8n-nodes-saidwire. Self-hosted:
npm install n8n-nodes-saidwireOpen WebUI
Voice input for Open WebUI through the OpenAI-compatible endpoint.
LiteLLM
Route transcription through a LiteLLM proxy to Saidwire.
Claude and Cursor via MCP
An MCP server that lets AI assistants such as Claude and Cursor transcribe audio with Saidwire.
pip install saidwire-mcpOpenAI-compatible
POST /v1/audio/transcriptions with the official OpenAI SDKs: change the base URL and the key.
Deepgram API-compatible
Streaming /v1/listen: change the base URL and the key in an existing client.
Listed for compatibility only. No partnership or endorsement is implied. "Coming" means not released yet.
Using another tool?
If it lets you set a custom base URL for OpenAI transcription or for Deepgram-style streaming, it should work. Tell us which one and we will write the guide.