specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Unreal Speech
providerId: unrealspeech
created: '2026-07-11'
modified: '2026-07-11'
reconciled: false
tags:
- Text to Speech
- TTS
- Rate Limiting
- Quotas
description: >-
  Unreal Speech enforces two kinds of limits. A per-second request-rate limit is
  tied to the account's plan tier (documented as roughly 2 requests/second on the
  free tier, 8 requests/second on Basic, and 32 requests/second on Pro-class
  plans). Separately, monthly consumption is metered by synthesized characters
  against the plan's included allowance, with overage or a plan upgrade required
  beyond it. Per-request payload is also bounded by endpoint: up to 1,000
  characters on /stream, up to 3,000 on /speech, and up to 500,000 on
  /synthesisTasks.
notes: >-
  The per-second request-rate figures reflect documented plan tiers at the time
  of review and may be adjusted per plan; confirm current values in the API
  reference and pricing page during reconciliation. Requests over the rate limit
  should be retried with exponential backoff.
sources:
- https://docs.unrealspeech.com/reference/getting-started-with-our-api
- https://app.unrealspeech.com/pricing
responseCodes:
  throttled: 429
limits:
- name: Free Tier Request Rate
  scope: account
  metric: requests
  limit: 2 per second
  notes: Documented request-rate ceiling on the free plan.
- name: Basic Tier Request Rate
  scope: account
  metric: requests
  limit: 8 per second
  notes: Documented request-rate ceiling on the Basic plan.
- name: Pro Tier Request Rate
  scope: account
  metric: requests
  limit: 32 per second
  notes: Documented request-rate ceiling on Pro-class plans.
- name: Monthly Character Allowance
  scope: account
  metric: characters
  limit: per plan
  notes: Synthesized characters are metered monthly against the plan's included allowance (free ~250K).
- name: Stream Payload Size
  scope: request
  metric: characters
  limit: 1000
  notes: Maximum characters per POST /stream request.
- name: Speech Payload Size
  scope: request
  metric: characters
  limit: 3000
  notes: Maximum characters per POST /speech request.
- name: Synthesis Task Payload Size
  scope: request
  metric: characters
  limit: 500000
  notes: Maximum characters per POST /synthesisTasks request.
policies:
- name: Backoff Strategy
  description: On a 429 response, retry with exponential backoff and jitter rather than hammering the endpoint.
- name: Endpoint Selection
  description: Choose the endpoint by text length - /stream for short low-latency text, /speech for medium text, /synthesisTasks for long-form - to stay within per-request payload limits.
- name: Plan Upgrade
  description: Sustained higher throughput or larger monthly volume is obtained by moving to a higher plan tier rather than a temporary limit increase.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com