Compare

AMDSure vs Twilio Answering Machine Detection

If your calls already go out through Twilio, its AMD is one parameter away. It only works on calls Twilio places, and the audio is already in Twilio's cloud.

AMDSureTwilio AMD
The 3% abandon ruleShadow mode measures people wrongly sent to voicemail on your own calls, beside your built-in AMD(), with the sample size: your detector's share of the 3% abandon rate, before you switch. People are decided first, 2.55 s on average.Published accuracy claims, no per-account error rate on people. Only on Twilio calls.
How it decidesA compiled agent on your Asterisk box measures the call (a real voice detector, tones, a learned acoustic model, the sound of the pickup itself); AMDSure's server applies the rules and signs the verdict.Twilio's engine analyses timing, pattern and frequency of the incoming audio on calls placed through its API.
Where the audio goesStays on your server. The agent sends numbers (speech probabilities, tone events, model scores), never audio, unless you switch learning on.The call is already on Twilio; the analysis happens there.
Time to a verdictMedian 4.6 s to a verdict on live calls from the pickup, 2.3 s on recordings that start mid-greeting; short greetings decide at about 2 s, beeps and known greetings sooner.Twilio's documentation: results within about 4 seconds of the answer on average with defaults; DetectMessageEnd waits for the greeting to end before reporting a machine.
PriceFree: 50,000 verdicts a month, two servers, no card. Starter $69 a month for 500,000 then $0.0002; Growth $269 for 5,000,000 then $0.00012; Scale $899 for 25,000,000 then $0.00008.Twilio's launch pricing was $0.0075 per call where enabled and answered; check Twilio's current pricing page.
What it reportsPerson, voicemail, call screener, unsure, each with a reason, a confidence and the time to the verdict. And what answered, by name: Verizon, AT&T and other carriers' default greetings, number read-outs, iPhone, Samsung and Google call screening, a disconnected number (the intercept tones), fax, busy, dead air, a false answer. Stamped on the call as AMDSURE_KIND for the dialer.AnsweredBy: human, machine_start, machine_end_beep, machine_end_silence, machine_end_other, fax, unknown.
Works onAsterisk 18 to 22 and everything built on it: VICIdial, ViciBox, FreePBX, Issabel, GoAutoDial, custom dialers. FreeSWITCH in development.Twilio Programmable Voice only.
Proof before you switchShadow mode runs beside your built-in AMD() on every call and stamps its verdict on the call; the dashboard shows people wrongly sent to voicemail and voicemails caught, against AMD(), before you switch.Published accuracy claims (94 percent for the enhanced engine at launch; near 100 percent for DetectMessageEnd in the US), no per-account side-by-side.
When it failsOpen, to your own AMD(). A brain outage costs seconds, never a call.Returns unknown.
Your audio and trainingYour call audio is used for training only with the learning checkbox on, which you can turn off at any time; phone numbers are never seen.Per Twilio's terms.

Where Twilio AMD is good

Built into Twilio's voice API. Per call.

  • Nothing to install when you are on Twilio.
  • Async mode connects the callee at once and reports in the background.

Where AMDSure differs

  • AMDSure works on your own Asterisk dialer and trunks, which Twilio's AMD cannot; Twilio's only works on Twilio calls.
  • At $0.0075 a call, 500,000 answered calls are $3,750 a month on Twilio against $69 on AMDSure Starter.
  • AMDSure names call screening assistants; Twilio reports them as human or unknown.

Sources

Facts about Twilio AMD come from its own documentation and pricing pages on the dates noted; tell us at [email protected] if something has changed and we will fix it. Other comparisons: Stock AMD(), FreeSWITCH AMD, AMDY, Telnyx AMD, Plivo, SignalWire AMD, Vonage, VM Hunter.

See it on your own traffic.

A 20 minute call. We show the console with live numbers, then set up shadow mode on your dialer so you can judge for yourself.

Start free