# podcast-transcriber

Subagent for turning podcast audio and video into structured JSON transcripts. It runs FFMPEG to process tracks, identifies speakers, and calculates exact timestamps.

- subagent · `type=agent` · Claude Code · 480 installs
- Author: davila7
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/ffmpeg-clip-team/podcast-transcriber.md
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/agents/podcast-transcriber

## What it does

This agent processes recordings directly through shell utilities. It starts by inspecting media streams with `ffprobe` to determine format details and runtime. Then it extracts the audio track into a single-channel 16 kHz WAV file via FFMPEG, applying a `loudnorm` filter whenever volume levels need leveling.

Longer audio runs are divided into chunks for easier handling. The agent assigns millisecond-level timestamps to each spoken phrase, distinguishes separate voices, and outputs everything as a JSON file. This payload includes confidence ratings alongside metadata like detected language and total run time.

Local `ffmpeg` and `ffprobe` binaries must be present in the path. Without them, shell execution fails immediately during the audio extraction stage.

## Install

### Claude Code (native platform)

1. **claude-code-templates**
```bash
npx claude-code-templates@latest --agent ffmpeg-clip-team/podcast-transcriber
```

2. **Manually: put the files in place**
- File: `.claude/agents/podcast-transcriber.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/podcast-transcriber.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/podcast-transcriber.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/ffmpeg-clip-team/podcast-transcriber.md

## When to use it
- extract audio from video files into 16 kHz mono WAV
- normalize uneven loudness across podcast recordings
- slice lengthy audio files into timed chunks
- generate JSON transcripts with speaker labels and millisecond timings

## Example prompts
- Extract the audio from `interview.mp4` and produce a JSON transcript with speaker tags.
- Normalize `episode_12.wav` and transcribe speech with millisecond timestamps.

## Similar
- [prompt-engineer](https://affpapa.org/en/nevibecode/agents/voltagent-community-prompt-engineer.md) — subagent · 16078 installs
- [prompt-engineer](https://affpapa.org/en/nevibecode/agents/prompt-engineer.md) — subagent · 16078 installs
- [ai-engineer](https://affpapa.org/en/nevibecode/agents/voltagent-community-ai-engineer.md) — subagent · 7687 installs
- [ai-engineer](https://affpapa.org/en/nevibecode/agents/ai-engineer.md) — subagent · 7687 installs
- [search-specialist](https://affpapa.org/en/nevibecode/agents/search-specialist.md) — subagent · 6886 installs
- [business-analyst](https://affpapa.org/en/nevibecode/agents/business-analyst.md) — subagent · 3855 installs

From the same set: [davila7/claude-code-templates](https://affpapa.org/en/nevibecode/by/davila7/claude-code-templates.md) — 804 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md