On-device transcription, forced alignment, sentence segmentation, karaoke captions, timed-transcript translation, and speaker naming through the local EdgeSpeak CLI. Requires edgespeak-cli on macOS Apple Silicon, Linux x86_64, or Windows x64.
Plugin edgespeak from marketplace edgespeak in lattifai/EdgeSpeak. Commands read from the repo's manifest — yesterday.
edgespeak-alignForce-align audio/video against a known transcript on-device via EdgeSpeak to produce word-level timestamps (start, end, score) for karaoke captions, word-accurate SRT, voice-over sync, and clip extraction. The default model covers 346 languages and mixed-language text; the transcript can be plain t
edgespeak-broadcastTurn text into natural speech fully on-device via EdgeSpeak (Broadcast) — synthesize WAV audio with official named voices, cloned voices, style instructions, speed and reproducible seeds, design a brand-new voice from a text description, and manage a local voice library including cloning a voice fro
edgespeak-karaokeCreate word-highlighted karaoke ASS subtitles, optionally bilingual with a translated second line, and burn them into local video using one EdgeSpeak transcription request with inline word-level forced alignment. Use when the user asks for karaoke captions, per-word highlighting, an ASS file, biling
edgespeak-name-speakersResolve anonymous speaker_N labels in an EdgeSpeak diarized transcript to evidence-backed real names, while preserving the original cluster IDs and leaving uncertain identities unresolved. Use after edgespeak-transcribe --diarize for interviews, meetings, panels, and podcasts when the user wants nam
edgespeak-segmentSplit a long run of text into natural sentences on-device via EdgeSpeak using a semantic sentence splitter that works on unpunctuated ASR output, or re-segment a word-timed transcript JSON (from transcribe/align) into new sentence boundaries while re-mapping every word timing. Use when the user has
edgespeak-transcribeTranscribe audio/video on-device via EdgeSpeak into text, JSON, or SRT, with optional word-level timing, anonymous speaker diarization (who said what), and sentence-shaping parameters for subtitles, meeting notes, voice memos, and searchable transcripts. Use when the user has a local media file to t
edgespeak-translateTranslate a timed transcript (EdgeSpeak JSON or SRT) into a target language yourself, keeping a strict 1:1 segment mapping and every timestamp untouched, so the result still lines up with the audio for subtitles and spoken versions. Enforces punctuation parity, consistent terminology, localized coun
edgespeak-video-voiceGive code-rendered video (HyperFrames, Remotion, Canvas, Manim, plain web animation) an on-device voice-over with word-synced captions — write the narration, synthesize it locally with EdgeSpeak, force-align it for word timestamps, and convert those into a cues.json / timing.js / frame-number table
edgespeak-yt-downloadDownload an authorized YouTube video's media, captions, and public metadata with yt-dlp for local transcription or analysis, using conservative serial requests, stable filenames, no-overwrite defaults, and safe cookie handling. Use when the user supplies a YouTube URL and needs a local audio/video f
On-device speech synthesis with EdgeSpeak Broadcast, voice-overs with word-synced captions for code-rendered video, and authorized YouTube media and metadata acquisition with yt-dlp. Companion to the edgespeak plugin.
9 skills
The diff pane as a plugin: /diff opens the session's uncommitted changes beside the transcript, file by file with their hunks, refreshed as Claude edits and runs commands; the first edit opens it where the terminal is wide enough, and a file's ask button rides its hunks on the next prompt.
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
A personal Claude Code impact report for justifying your usage to a manager or a self-review: what you shipped, which projects it went to, and each project's share of your usage. Reads your ~/.claude/projects transcripts and runs read-only git locally; only counts and project names are sent to write it up, and the report stays on your machine.
1 skill
Local session memory for coding agents: indexes existing session transcripts on disk across 43 coding agents into a local BM25 store with MCP recall, hooks, skill, and /deja command
2 skills · 1 command · 1 MCP server · hooks
Plan, build, and debug Zoom integrations across REST APIs, Meeting SDK, Video SDK, webhooks, bots, and MCP workflows. Search meetings, retrieve recordings, access transcripts, and design AI-powered Zoom experiences.
44 skills · 3 MCP servers
Answer questions about past agent runs from the recording, not from memory. Three read-only stdlib scripts over the session transcripts Claude Code and Codex already write: a finder that locates the run by repo, time window, touched file or command text; a timeline that numbers every step with its exit code, lists every file changed (flagging changes the agent never mentioned) and walks back from any step with each causal edge labelled recorded or inferred by a named rule; and a claim checker that tests a summary's claims (ran, passed, edited, not-edited, read) against the run and catches tests that passed before a later edit and exit codes that belong to the last command of a pipe. Secrets are always redacted. A UserPromptSubmit hook gives the agent this session's transcript path when a prompt asks about past work. Ships cs-run-forensics agent, /cs-run-forensics and /cs-check-claims commands and 3 cited references.
1 skill · 2 commands · 1 agent · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.