Build speech features with AssemblyAI's APIs.
Guides implementation of AssemblyAI speech features using API and SDK patterns that avoid deprecated interfaces and authentication mistakes.
When to use it
Use when implementing transcription, streaming speech recognition, dictation, audio intelligence, or voice agents with AssemblyAI, including LiveKit and Pipecat integrations.
It changes how the agent implements AssemblyAI speech features, including checking installed SDK versions before writing integration code.
What you provide
The Node SDK example reads the AssemblyAI API key from ASSEMBLYAI_API_KEY to authenticate the client.
No additional actions listed.