# Audio Voice Activity Segmentation Detect likely speech-range activity and export speech/non-speech ranges as JSON, CSV, SRT markers, or an FFmpeg cut list. > Canonical page: https://elysiatools.com/en/tools/audio-voice-activity-segmentation - **Category:** Media - **Keywords:** voice activity detection, vad, speech segmentation, speech timeline, silence detection, srt, ffmpeg cut list ## Overview This deterministic VAD analyzes window energy, speech-band energy, and zero-crossing behavior to separate likely speech activity from non-speech. It is useful for timing, trim preparation, and subtitle alignment, but it does not identify words, speakers, or language and cannot prove that a sound is human speech. Export formats include full JSON/CSV timelines, SRT activity markers, and an FFmpeg-ready command list for the detected active ranges. ## Inputs - **Audio File** (file): Select audio to segment - **Sensitivity** (select) - **Minimum Speech (s)** (number) - **Minimum Silence (s)** (number) - **Export Format** (select) ## When to use - When preparing long audio recordings for editing by identifying and removing silent gaps or non-speech segments. - When generating rough subtitle timing markers (SRT) to align speech segments before transcription. - When creating automated FFmpeg cut lists to programmatically split or trim audio files based on voice activity. ## How it works - Upload your audio file and select the detection sensitivity level (conservative, balanced, or sensitive). - Define the minimum duration thresholds for speech and silence segments to filter out brief noises or short pauses. - Choose your preferred export format, such as a JSON/CSV timeline, SRT markers, or an FFmpeg cut list. - Run the analysis to process the audio's energy and zero-crossing behavior, then download the generated segment file. ## Use cases - Pre-processing podcast recordings to generate an FFmpeg cut list that trims out long pauses and dead air. - Generating SRT activity markers to establish baseline timestamps for manual subtitle alignment. - Exporting a JSON timeline of speech segments to feed into custom automated audio processing pipelines. ## Frequently asked questions ### Does this tool transcribe the speech or identify the speaker? No, this tool only detects the presence of speech-range audio activity; it does not perform speech-to-text transcription or speaker identification. ### What does the FFmpeg cut list export format do? It generates a list of time-based commands that you can run directly in FFmpeg to split or trim the audio file at the detected speech boundaries. ### How do the sensitivity settings affect the segmentation? Conservative mode requires stronger speech signals to trigger detection, while sensitive mode captures quieter speech but may include more background noise. ### Can I adjust how short pauses are handled? Yes, you can configure the minimum silence duration to prevent short pauses between words from splitting a single speech segment. ### What audio file formats are supported? The tool accepts standard audio file formats up to 100 MB for voice activity analysis. ## Related tools - [Audio Chord Progression Detector](https://elysiatools.com/en/tools/audio-chord-progression-detector): Estimate an audio file's chord sequence locally and download CSV, JSON, SVG, and method notes in one ZIP. - [Audio Silence Map](https://elysiatools.com/en/tools/audio-silence-map): Output silence intervals as JSON and CSV - [Dataset Imbalance Detector & Resampler](https://elysiatools.com/en/tools/dataset-imbalance-detector-resampler): Detect class imbalance in CSV or JSON datasets, compare resampling strategies, and preview a balanced output dataset - [JSON Path Visualizer](https://elysiatools.com/en/tools/json-path-visualizer): Visualize JSON or JSONL as an expandable tree and copy JSONPath expressions for every node - [Mock Data Prefix / Abbreviation Conflict Detector](https://elysiatools.com/en/tools/mock-data-naming-conflict-detector): Detect visually confusing field names and prefix collisions in CSV, JSON, or schema inputs, then suggest clearer renames - [Time Series Anomaly Detector](https://elysiatools.com/en/tools/time-series-anomaly-detector): Upload CSV or JSON time series data, detect anomalies with Z-Score and IQR methods, and return a chart-backed report - [Fitness Workout PDF Generator](https://elysiatools.com/en/tools/fitness-workout-pdf-generator): Generate a printable workout-plan PDF from JSON or CSV data with exercise tables, progress boxes, and optional QR video links - [PDF Table Extractor to CSV/JSON](https://elysiatools.com/en/tools/pdf-table-extractor-to-csv-json): Extract tables from PDFs with OpenDataLoader and export them as structured JSON, flat CSV, or HTML tables ## Samples - [Copyright-Free FLAC Audio Samples](https://elysiatools.com/en/samples/flac-samples): Lossless FLAC audio samples for testing and development, mirrored from MP3 set with nature sounds and meditation music - [Copyright-Free MP3 Audio Samples](https://elysiatools.com/en/samples/mp3-samples): Collection of royalty-free audio samples for testing and development purposes including nature sounds, meditation music, and ambient audio - [Copyright-Free WAV Audio Samples](https://elysiatools.com/en/samples/wav-samples): Uncompressed PCM WAV audio samples for testing and development, mirrored from MP3 set with nature sounds and meditation music - [Speech Learning & Safety Audio Samples](https://elysiatools.com/en/samples/audio-learning-safety-samples): Deterministic synthetic WAV inputs for pronunciation comparison, dictation preflight, alert degradation, and voice privacy tools.