SKILL.md

speech-to-text

Agent skill frombenchflow-ai/skillsbench

Transcribe video to timestamped text using Whisper tiny model (pre-installed).

install speech-to-text

# install the skillsbench pack with the skills CLI

npx skills add benchflow-ai/skillsbench

# or add only this skill to Claude Code: copy the folder

tasks-extra/video-tutorial-indexer/environment/skills/speech-to-text/

# into

~/.claude/skills/speech-to-text/ (all projects)

.claude/skills/speech-to-text/ (one project)

tasks-extra/video-tutorial-indexer/environment/skills/speech-to-text/SKILL.md

Apache-2.0Full file

1# Speech-to-Text
2 
3Transcribe video to text with timestamps.
4 
5## Usage
6 
7```bash
8python3 scripts/transcribe.py /root/tutorial_video.mp4 -o transcript.txt --model tiny
9```
10 
11This produces output like:
12```
13[0.0s - 5.2s] Welcome to this tutorial.
14[5.2s - 12.8s] Today we're going to learn...
15```
16 
17The tiny model is pre-downloaded and takes ~2 minutes for a 23-min video.

Excerpt shown under the repository's Apache-2.0 license.

How to install speech-to-text

tasks-extra/video-tutorial-indexer/environment/skills/speech-to-text/SKILL.md

More skills in skillsbench

Newsletter

Join the Community