Wispr Flow语音转文本参考架构

v20260423

wispr-reference-architecture

本指南提供了Wispr Flow语音转文本API的参考架构和最佳实践。它详细介绍了如何使用WebSocket和REST API进行语音转文本集成，涵盖了音频格式要求、上下文感知（尤其针对代码编程场景）以及全面的错误处理策略。

语音识别 API 集成语音转文本流式 WebSocket SaaS

370 次下载

概览

Wispr Flow Reference Architecture

Guidance for reference architecture with Wispr Flow voice-to-text API.

WebSocket API: wss://api.wisprflow.ai/api/v1/ws (recommended, low latency)
REST API: POST /api/v1/transcribe (simpler, higher latency)
Auth: API key (backend) or access token (client-side)
Audio format: 16kHz mono PCM preferred
Context awareness: Understands code, CLI commands, dev jargon
Platforms: Mac, Windows, iOS, browser API

Error	Cause	Solution
`401 Unauthorized`	Invalid key	Check at wisprflow.ai/developers
WebSocket closed	Network issue	Reconnect with backoff
Poor accuracy	Wrong context	Set context to 'programming' for code

See related Wispr Flow skills for more patterns.

信息

分类编程开发

名称 wispr-reference-architecture

版本 v20260423

大小 1.54KB

来源 jeremylongshore/claude-code-plugins-plus-skills

更新时间 2026-04-28