# Real-Time Voicemail Detection in Telephony Audio Using Temporal Speech Activity Features

Source: [Real-Time Voicemail Detection in Telephony Audio Using Temporal Speech Activity Features](<https://www.sota2.com/research/paper/69ddca047d0172140ec79925>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-15T05:00:04.591Z

Authors: Kumar Saurav

Published: 2026-04-02T17:43:24.000Z

## Abstract

Outbound AI calling systems must distinguish voicemail greetings from live human answers in real time to avoid wasted agent interactions and dropped calls. We present a lightweight approach that extracts 15 temporal features from the speech activity pattern of a pre-trained neural voice activity detector (VAD), then classifies with a shallow tree-based ensemble. Across two evaluation sets totaling 764 telephony recordings, the system achieves a combined 96.1% accuracy (734/764), with 99.3% (139/140) on an expert-labeled test set and 95.4% (595/624) on a held-out production set. In production validation over 77,000 calls, it maintained a 0.3% false positive rate and 1.3% false negative rate. End-to-end inference completes in 46 ms on a commodity dual-core CPU with no GPU, supporting 380+ concurrent WebSocket calls. In our search over 3,780 model, feature, and threshold combinations, feature importance was concentrated in three temporal variables. Adding transcription keywords or beep-based features did not improve the best real-time configuration and increased latency substantially. Our results suggest that temporal speech patterns are a strong signal for distinguishing voicemail greetings from live human answers.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Voicemail detection on 77,000 Outbound Calls (production)](<https://www.sota2.com/research/sota/voicemail-detection-on-77-000-outbound-calls-production.md>): Voicemail detection; 77,000 Outbound Calls (production); False Positive Rate (FPR): 30.

- [Answering Machine Detection on Production (val)](<https://www.sota2.com/research/sota/answering-machine-detection-on-production-val.md>): Answering Machine Detection; Production (val); Accuracy: 96.1.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.09675>)

[PDF](<https://arxiv.org/pdf/2604.09675.pdf>)