逸尘自用的唯一音视频转写入口，统一管理 StepFun Step ASR、火山引擎豆包 ASR 与本地 Whisper 三个后端：纯文本转写按参数、已保存的引擎偏好、默认 StepFun 的顺序选择引擎并处理首次选择提示；时间戳、SRT 字幕与口播粗剪转写在 Step 国内站与豆包之间按输出需求、安全边界和可用状态路由；另含转写前体检。适用于本地音频或视频，包括小宇宙、B 站等已下载的音视频、会议录音和长音频重新识别。用户说用 Step 转写、只用 step、stepaudio-2.5-asr、用豆包或火山转写、用本地 Whisper 免费转写、设置转写引擎，或点名 step-asr、volc-asr、yichen-asr 时使用；用户明确指定服务商时不得静默切换。Use when a local audio or video file needs transcription, timestamps or subtitles, or when the user names $yichen-asr, $step-asr or $volc-asr.
Install
KitScout test notes
Not tested by KitScout yet. When our harness runs it, pros, cons and the tasks we tried appear here, with the date and evidence.
Similar tools
audio-jingle107,550 ★ Audio generation skill — jingles, beds, voiceover, and sound effects. Routes music requests to Suno V5 / Udio
deepgram-automation79,451 ★ Automate Deepgram tasks via Rube MCP (Composio). Always search tools first for current schemas.
ElevenLabs Automation79,451 ★ Automate ElevenLabs text-to-speech workflows -- generate speech from text, browse and inspect voices, check su