| Name | Description | Recommended value |
|---|---|---|
| maximum word length (ms) | Maximum duration of a single Voice utterance allowed. | 5000 |
| maximum number of words | Maximum number of words in the greeting If exceeded, then the result is detection as a MACHINE | 3 |
| between words silence (ms) | Minimum duration of silence after a word to consider the audio what follows as a new word | 50 |
| minimum word length | Minimum duration of Voice to considered as a word | 100 |
| silence threshold (ms) | If the average level of noise in a sample does not reach this value, from a scale of 0 to 32767, then we will consider it to be silence. | 256 |
| total analysis time (ms) | Maximum time allowed for the algorithm to decide on whether the audio represents a HUMAN, or a MACHINE | 5500 |
| after greetings silence (ms) | Silence after detecting a greeting. If exceeded, then the result is detection as a HUMAN | 800 |
| greeting (ms) | Maximum length of a greeting. If exceeded, then the result is detection as a MACHINE. | 1500 |
| Initial silence (ms) | Maximum silence duration before the greeting. If exceeded then MACHINE. | 5100 |
Notes on Answering Machine Detection
- AMD can only be used in Voice Broadcasting and Predictive dialing modes.
- AMD will not detect 100% of the answering machines.
- Calls will take longer to get to agents because the system listens to the first few seconds of the call to determine if it is an Answering Machine or a live person.