腾讯云语音识别 ASR Skill,适用于语音转文字、音频转写、字幕生成、会议转录、语音消息识别、 本地文件或 URL 音频识别。包含三种模式:一句话识别(<=60s 短音频)、录音识别极速版 (<=2h/100MB 中长音频快速同步返回)、录音识别(<=5h 长音频异步识别)。支持普通话、 英语、粤语、日语、韩语、德语等语种,以及中英粤混说和多种中文方言。
Version 0.1.4 - Added self-diagnostic script (`scripts/self_check.py`) to validate Tencent Cloud credentials and ASR capability before real recognition. - Introduced new `sentence_recognize.py` script, replacing the previous `main.py` for short audio recognition. - Added comprehensive, step-by-step activation and configuration guides in references (e.g., `tencent_cloud_activation.md`, `env_config.md`). - Included sample audio (`assets/16k.wav`) and visual guides for easier onboarding. - Updated documentation for clearer workflow, routing rules, engine options, and common troubleshooting.