Question
- Why does it take a long time before my NCCO is executed?
- Why do my end-users experience 3-5 seconds of silence at the start of the call?
Applies To
- Voice API
- Machine Detection
Answer
Customers using the standard machine detection feature with the Vonage Voice API might notice a 3-5 second period of silence at the start of a call. This brief silence allows Vonage's machine detection system to analyze the audio and determine whether a human or a machine has answered the call.
To reduce this silence at the beginning of calls, consider the following alternatives:
-
Switch to Advanced Machine Detection in Asynchronous Mode
Advanced Machine Detection in asynchronous mode(using "mode": “default” only) allows the machine detection process to occur concurrently with NCCO processing. This parallel processing reduces the extended silence experienced and enhances the overall call experience.Note: Advanced Machine Detection is a premium feature with an additional charge of €0.0070 per call. For more details, refer to our developer documentation.
-
Disable Machine Detection
If distinguishing between human and machine responses is not crucial for your application, you can disable machine detection. This will reduce the extended silence and allow immediate processing of NCCO instructions, resulting in a smoother call experience.
Disabling machine detection is ideal when quick response times are essential, and there is no need to handle machine responses differently from human responses.
Additional Information
For more information on machine detection, refer to the following resources:
Related to:
Articles in this section
- Why is there an extended silence at the beginning of a call and how can it be reduced?
- Intermittent WebSocket Disconnections
- What Is An NCCO
- How Does Jitter Impact Voice Calls
- Voice API "Bad Request" Error When Ending Calls Immediately After Creation
- What Is MOS Score In Voice Calls?
- Is there a way to see current active calls on our account?
- How To Make MP3 of Recording From Voice API in Python
- How To Enter DTMF Before MP3 Stream Finishes
- Can a user call leg be removed from the voice call recording?