Increased latency and API errors
minor- Started
- Nov 18, 10:34 AM EST
- Resolved
- Nov 18, 4:02 PM EST
- Duration
- 5h 28m
ElevenLabs's updates (8)
- investigating · Nov 18, 10:34 AM EST
We’re seeing an increase in API errors and latency for multiple products. Agent conversations and dub creation are not impacted by this issue. We’re currently investigating and will post updates as soon as they become available.
- investigating · Nov 18, 10:41 AM EST
We’re now seeing that some agent conversations are impacted by this issue. We’re still investigating this and will continue to post updates as they become available.
- investigating · Nov 18, 12:20 PM EST
We are still investigating issues impacting services in the US residency. Products with confirmed impact: - TTS — partial elevated latency and some failing requests - STT — partial elevated latency and some failing requests - ElevenAgents — further updates pending - Dubbing — elevated failure rate - Music — partial elevated latency and failing requests - Creative — issues loading UI
- investigating · Nov 18, 12:33 PM EST
We are still investigating issues impacting services in the US residency. Update on ElevenAgents: we can confirm that outbound telephony calls are operational, and quality within an active conversation is unaffected. Affected with degraded performance / partial failure: - some inbound websocket calls fail to connect - some inbound telephony calls (including Twilio and SIP) fail to connect - increased latency and elevated failure rates in agent get, update, create, delete operations - increased slowness in UI operations
- identified · Nov 18, 1:36 PM EST
We’ve identified the cause of the ongoing US residency degradation and have begun stabilizing the underlying systems. Median response times have partially recovered, and some users may still experience intermittent errors and elevated latency across the following products: - TTS, STT & Music — partial elevated latency and some failing requests - Voices & API — elevated latency and intermittent errors when fetching or loading voices - Dubbing — elevated failure rate and delays (showing early signs of improvement) - Creative & Dashboard UI — slowness and issues loading pages - ElevenAgents — outbound telephony and active conversation quality remain operational; some inbound telephony (SIP/Twilio) and WebSocket calls may fail to connect, and agent management/UI operations remain slow We are actively working to restore full functionality, and rolling out fixes to permanently reduce pressure on our systems. We will share further updates when these mitigation steps are rolled out.
- identified · Nov 18, 2:33 PM EST
We’re still working to fully restore functionality and address the root cause for this issue. We will continue posting updates about this as they become available.
- monitoring · Nov 18, 3:32 PM EST
After further observation, latency is back to normal levels and we’re seeing a reduction in error rates across all products. We’re currently still monitoring to confirm resolution.
- resolved · Nov 18, 4:02 PM EST
Systems are now operational across all products. We will post a root cause analysis after an extensive review.
What our checks recorded
We checked ElevenLabs 66 times from ElevenLabs's start time to its resolution, once every five minutes. Every one of them reached ElevenLabs's site, so this incident did not show up as a failed page load. Our check loads the public site; it cannot see an API error or a broken feature. Median response time was 641 ms, against 648 ms over the 24 hours before. ElevenLabs posted the incident at Nov 18, 10:34 AM EST.
Updates are ElevenLabs's own words from its status page (status: resolved); we never add an inferred cause.