Transcribes audio/video to text, generates SRT subtitles, and performs speaker segmentation using Whisper/faster-whisper. Useful for meeting recordings, voice memos, and interviews.
How this skill is triggered — by the user, by Claude, or both
Slash command
/sunflowermm-xrk-agt-2:office-transcribeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
会议录音、语音备忘、采访、`.mp3` `.wav` `.m4a` 转文字、要 SRT 字幕。
会议录音、语音备忘、采访、.mp3 .wav .m4a 转文字、要 SRT 字幕。
run 执行转写脚本,输出 .txt / .srt / .jsonoffice-meeting 出纪要;短备忘:摘要 5 条from faster_whisper import WhisperModel
model = WhisperModel("large-v3", device="cpu", compute_type="int8")
segments, info = model.transcribe("meeting.mp3", language="zh", vad_filter=True)
lines = []
for seg in segments:
lines.append(f"[{seg.start:.1f}s-{seg.end:.1f}s] {seg.text.strip()}")
open("transcript.txt", "w", encoding="utf-8").write("\n".join(lines))
pip install faster-whisper
# 可选 GPU:device=cuda
Whisper 原生无 diarization;需 pyannote 等额外模型时先说明环境与授权。
def fmt(t):
h, r = divmod(int(t), 3600); m, s = divmod(r, 60); ms = int((t % 1) * 1000)
return f"{h:02d}:{m:02d}:{s:02d},{ms:03d}"
# 每 segment 写一条 SRT 块
memory/无 faster-whisper / 无 run → 请用户提供文字稿或外部转写结果;见 office-env-setup
npx claudepluginhub joshuarweaver/cascade-code-general-misc-4 --plugin sunflowermm-xrk-agt-2Transcribes local audio/video files to Markdown and SRT subtitles using MLX/Whisper (Apple Silicon) or faster-whisper (others).
Transcribes audio/video to speaker-labeled text with local MLX (Apple Silicon) or remote API. Also handles ASR preprocessing: format conversion, 16kHz mono downsampling, multi-segment merging, and M4A transcoding with pitch-preserved speedup.
Transcribe audio files to text with optional diarization and known-speaker hints. Use when a user asks to transcribe speech from audio/video, extract text from recordings, or label speakers in interviews or meetings.