Skip to content
Voice Monkey Docs
Heads up — these docs cover Voice Monkey API v3, the current version. If you signed up before the v3 launch, your account is still on API v2 for a limited transition period and the examples below will not work against your account.

API

The Voice Monkey API lets you trigger Alexa Routines, make text-to-speech announcements, start Flows and read/write variables with a simple HTTP request from anywhere.

Endpoints

Section titled “Endpoints”

All endpoints live under:

https://api-v3.voicemonkey.io
EndpointPurpose
GET/POST /announcePlay a text-to-speech announcement (and optional media) on a Speaker device.
GET/POST /triggerTrigger a Routine Trigger device (fires its Alexa Routine).
GET/POST /flowStart a Flow by its request reference.
GET/POST /devicesList the Speakers and Routine Triggers on the authenticated account.
GET/POST/PUT/DELETE /variablesRead, create, update or delete variables.
POST/PUT /catch/{userId}/{webhookToken}Receive an inbound webhook payload.

See the per-endpoint docs for the full parameter list:

Every authenticated endpoint accepts a var-<NAME>=value shorthand parameter that upserts a variable as a side effect of the call — see Variables API → Setting variables alongside any other API call.

Versioning

Section titled “Versioning”

The Voice Monkey API is currently at v3. All documentation on this site refers exclusively to v3.

  • v3 — current. Hosted at https://api-v3.voicemonkey.io.
  • v1 and v2 — deprecated. Existing integrations against api-v2.voicemonkey.io continue to work for now, but are no longer publicly documented and will not receive new features. New integrations should target v3.

Migrating from v2 to v3 is largely a hostname swap (api-v2 → api-v3) plus a handful of small parameter changes — most notably the announcement endpoint moves from /announcement to /announce and its TTS field is now called speech (was text). See the per-endpoint pages for the exact v3 contract.

Rate limits

Section titled “Rate limits”

Every account is bounded by two independent layers:

  • Plan quota — a monthly request budget set by your plan (Free 200/mo, Hobby 15,000/mo, Ultimate unlimited). Hitting the cap returns a MONTHLY_QUOTA_EXCEEDED error until the next billing period.
  • Safety throttle — a per-user burst limit that protects Alexa from runaway traffic. Tripping it triggers a short lockout during which calls return 429 Too Many Requests with a THROTTLED:<until> body. Independent of plan quota.

Always implement an exponential backoff on 429 responses.