overcast-voiceprint

overcast-voiceprint is a skill for Claude Code from kdr/overcast. It costs 55 tokens per session (927 once invoked), scanned A, original, Apache-2.0.

A local voice-recognition tool that compares recordings with a reference speaker and finds where that speaker talks in audio or video. A voice print is a numerical representation of voice characteristics, not a phrase transcript.

In plain words
What is it for?
Use it to enroll reference recordings, scan clips for a matching speaker, rank the match from 0 to 100, and identify recordings containing that speaker.
Why use it?
It helps check whether the same speaker appears in recordings and locate their sections without sending the media away from the case.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the overcast plugin — 35 skills, 1 hook shipped together

Good fit Use it to enroll reference recordings, scan clips for a matching speaker, rank the match from 0 to 100, and identify recordings containing that speaker.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kdr/overcast/overcast-voiceprint
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add kdr/overcast --skill overcast-voiceprint
Clone the repo
git clone --depth 1 https://github.com/kdr/overcast

Made for: Claude Code.

Or install overcast, the plugin that ships this one along with the rest of its 35 skills, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for overcast-voiceprint

README.md
[![agentmods](https://agentmods.dev/badge/skills/kdr/overcast/overcast-voiceprint.svg)](https://agentmods.dev/skills/kdr/overcast/overcast-voiceprint)
Your own site
<a href="https://agentmods.dev/skills/kdr/overcast/overcast-voiceprint"><img src="https://agentmods.dev/badge/skills/kdr/overcast/overcast-voiceprint.svg" alt="Measured on agentmods" height="20"></a>
Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 927 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00055 $0.00927
Opus 5 $0.00028 $0.00464
Sonnet 5 $0.00011 $0.00185
Haiku 4.5 $0.00006 $0.00093

Measured 8d ago against content hash 194d1f4d38b7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

overcast-voiceprint scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/overcast-voiceprint/SKILL.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

overcast-voiceprint

Use this skill to answer "is this the same speaker, and where do they talk?" with the local voice-print DB (pyannote/wespeaker speaker embeddings — no media leaves the case, ungated windowed default). It verifies a VOICE, not a phrase. Use the broad overcast skill and overcast/reference/verbs.md for exact flags.

Prerequisites

overcast doctor --json                 # confirm uv + visual-db (pyannote) are ready
scripts/visual-db-uv.sh --voice        # install pyannote.audio (once per machine)
overcast case init --json
overcast index create voices --type voice-print --local --json

Workflow

  1. Enroll the reference speaker (and any known recordings). voice add embeds the clip's voiced windows into the index:
overcast voice add ./known-speaker.wav --index voices --json     # the reference person
overcast voice add ./interview.m4a --index voices --json          # other recordings to search
  1. Rank WHERE the reference speaker talks inside a clip (pairwise, windowed scan). voice match <clip> <sample> scans <clip> for the <sample> speaker; --diarize upgrades to overlap-aware diarize-then-match (HF_TOKEN + accepted pyannote license, windowed fallback if ungated):
overcast voice match ./call.wav ./known-speaker.wav --json                 # where in call.wav does the sample speaker talk?
overcast voice match ./call.wav ./known-speaker.wav --diarize --min-margin 10 --json  # overlap-aware; gate on margin
  1. Rank WHICH enrolled recordings contain the reference speaker (index search):
overcast voice match ./known-speaker.wav --index voices --json   # members ranked by the speaker's presence
  1. Read the score honestly and corroborate before concluding. similarity is an anchored-cosine 0–100 RANK score (never 0–1) plus a raw cosine; --min-margin gates best-vs-runner-up. Corroborate with content (a listen transcript) before naming anyone, then promote through triage:

Read the full file on GitHub · 81 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 81 lines · 55 tokens per session scan A 194d1f4d38b7

Subscribe to this mod's changes

overcast-voiceprint is a skill published in the GitHub repository kdr/overcast (16 stars, last pushed 5d ago), licensed Apache-2.0. It adds 55 tokens to every session and 927 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.