Product details change. This page was checked against Wispr Flow's official website or documentation on 2026-08-17.
Wispr provides a Flow API for developers who want its speech processing inside their own application. Access is currently approval-based rather than open self-service.
Get access
The official quickstart says an organization must be approved before using the developer dashboard. Approved users sign in at platform.wisprflow.ai, open API Keys, and create a key. The key is displayed once, so store it in a secret manager rather than source code.
Audio requirements
The quickstart currently specifies base64-encoded, 16 kHz PCM WAV audio. Browser recording commonly produces WebM instead, so web applications need to convert the recording before sending it.
That conversion is an important architectural constraint: validate duration, format, and payload size before calling the API, and keep keys on a trusted backend. Never expose a production API key in client-side JavaScript.
API versus the Flow app
The consumer app provides ready-made dictation, shortcuts, personal vocabulary, and insertion into other apps. The API provides a building block. Choosing it means you own recording, format conversion, user consent, retries, storage, and the interface around the result.
Use the current official API documentation for endpoints and request examples. Exclusive-access terms and technical requirements can change faster than an independent article.