Whisper Local API - Secure & Private ASR
Deploy a privacy-first, 100% local speech-to-text service in a deterministic way. This allows OpenClaw to process audio transcriptions safely on your own hardware without ever contacting third-party cloud APIs.
Key SEO & Security Features
- * 100% Offline & Private: Your voice data, commands, and transcriptions never leave your host system. Zero cloud dependencies.
- Highly Accurate: Uses the
large-v3-turbo model via faster-whisper, achieving state-of-the-art accuracy even with accents or background noise. - Memory Safe: Operates around ~400-500MB of RAM, making it extremely lightweight for VPS or low-resource edge servers.
- OpenAI API Compatible: Exposes a strict
/v1/audio/transcriptions endpoint mimicking OpenAI's JSON format. Compatible natively with any software that supports OpenAI's Whisper API.
Standard Workflow
- 1. Install/update runtime:
bash scripts/bootstrap.sh
- 2. Start service:
bash scripts/start.sh
- 3. Validate service health:
bash scripts/healthcheck.sh
- 4. (Optional) Run a smoke transcription test with a local audio file:
CODEBLOCK3
Repo Location
Default install/update path used by scripts:
Override with env var before running scripts:
CODEBLOCK4
OpenClaw Integration Notes
After the healthcheck passes, use the secure local endpoint:
- * URL: INLINECODE4
- Endpoint: INLINECODE5
No authentication tokens are passed over the network.
Safety Rules
- * Ask before any package-manager operations.
- The API securely binds locally to
0.0.0.0. If exposing to the public internet, deploy behind a secure reverse proxy (like Nginx) and enforce HTTPS + Basic Auth. - This service will safely auto-fallback memory allocation modes (
float16 -> int8) to prevent CPU crashes.
Whisper Local API - 安全且私密的语音识别服务
以确定性方式部署一个隐私优先、100%本地的语音转文字服务。这使得OpenClaw能够在您自己的硬件上安全地处理音频转录，无需接触任何第三方云API。
核心SEO与安全特性
- * 100%离线且私密： 您的语音数据、指令和转录内容永远不会离开主机系统。零云依赖。
- 高精度识别： 通过faster-whisper使用large-v3-turbo模型，即使在带口音或背景噪音的环境下也能实现最先进的识别精度。
- 内存安全： 运行内存占用约400-500MB，非常适合VPS或低资源边缘服务器。
- 兼容OpenAI API： 暴露严格的/v1/audio/transcriptions端点，模拟OpenAI的JSON格式。原生兼容任何支持OpenAI Whisper API的软件。
标准工作流程
- 1. 安装/更新运行时：
bash
bash scripts/bootstrap.sh
- 2. 启动服务：
bash
bash scripts/start.sh
- 3. 验证服务健康状态：
bash
bash scripts/healthcheck.sh
- 4. （可选）使用本地音频文件进行烟雾转录测试：
bash
bash scripts/smoke-test.sh /path/to/test-speech.mp3
仓库位置
脚本使用的默认安装/更新路径：
在运行脚本前可通过环境变量覆盖：
bash
WHISPER_DIR=/custom/path bash scripts/bootstrap.sh
OpenClaw集成说明
健康检查通过后，使用安全的本地端点：
- * 地址：http://localhost:9000
- 端点：/v1/audio/transcriptions
无需通过网络传递任何认证令牌。
安全规则
- * 执行任何包管理器操作前需征得确认。
- API安全地本地绑定到0.0.0.0。如需向公共互联网开放，请部署在安全的反向代理（如Nginx）之后，并强制启用HTTPS和基本认证。
- 本服务将安全地自动回退内存分配模式（float16 -> int8），以防止CPU崩溃。
