Skillsbench pause-detector

Detect pauses and silence in audio using local dynamic thresholds. Use when you need to find natural pauses in lectures, board-writing silences, or breaks between sections. Uses local context comparison to avoid false positives from volume variation.

install

source · Clone the upstream repo

git clone https://github.com/benchflow-ai/skillsbench

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/benchflow-ai/skillsbench "$T" && mkdir -p ~/.claude/skills && cp -r "$T/tasks/video-silence-remover/environment/skills/pause-detector" ~/.claude/skills/benchflow-ai-skillsbench-pause-detector && rm -rf "$T"

manifest: tasks/video-silence-remover/environment/skills/pause-detector/SKILL.md

source content

Pause Detector

Detects pauses and low-energy segments using local dynamic thresholds on pre-computed energy data. Unlike global threshold methods, this compares energy to surrounding context, avoiding false positives when speaker volume varies.

Use Cases

Detecting natural pauses in lectures
Finding board-writing silences
Identifying breaks between sections

Usage

python3 /root/.claude/skills/pause-detector/scripts/detect_pauses.py \
    --energies /path/to/energies.json \
    --output /path/to/pauses.json

Parameters

```
--energies
```
: Path to energy JSON file (from energy-calculator)
```
--output
```
: Path to output JSON file
```
--start-time
```
: Start analyzing from this second (default: 0)
```
--threshold-ratio
```
: Ratio of local average for low energy (default: 0.5)
```
--min-duration
```
: Minimum pause duration in seconds (default: 2)
```
--window-size
```
: Local average window size (default: 30)

Output Format

{
  "method": "local_dynamic_threshold",
  "segments": [
    {"start": 610, "end": 613, "duration": 3},
    {"start": 720, "end": 724, "duration": 4}
  ],
  "total_segments": 11,
  "total_duration_seconds": 28,
  "parameters": {
    "threshold_ratio": 0.5,
    "window_size": 30,
    "min_duration": 2
  }
}

How It Works

Load pre-computed energy data
Calculate local average using sliding window
Mark seconds where energy < local_avg × threshold_ratio
Group consecutive low-energy seconds into segments
Filter segments by minimum duration

Dependencies

Python 3.11+
numpy
scipy

Example

# Detect pauses after opening (start at 221s)
python3 /root/.claude/skills/pause-detector/scripts/detect_pauses.py \
    --energies energies.json \
    --start-time 221 \
    --output pauses.json

# Result: Found 11 pauses totaling 28 seconds

Parameters Tuning

Parameter	Lower Value	Higher Value
`threshold_ratio`	More aggressive	More conservative
`min_duration`	Shorter pauses	Longer pauses only
`window_size`	Local context	Broader context

Notes

Requires energy data from energy-calculator skill
Use --start-time to skip detected opening
Local threshold adapts to varying speaker volume