ElevenLabs Outage History

Status: Identified We have identified the root cause of the issue and are currently preparing a release to address it. Affected components UI (Degraded performance)

Status: Identified Requests made to the voice "George", ID: JBFqnCBsd6RMkjVDRZzb, started failing at 19:30 UTC today. We have identified the cause of the issue and are working to release a fix for this. As a temporary solution, adding the voice ID 6WwXjDDEMyNmFG95zycZ from the library will allow generations that were previously directed at the voice of George to complete. The voice of George will soon be accessible again. If you are making requests to legacy voices "legacy voices": https://help.elevenlabs.io/hc/en-us/articles/26928417254801-What-are-Legacy-voices and getting the same error, please switch to a non-legacy voice. Affected components Text to Speech (Degraded performance) Conversations (Degraded performance)

Status: Identified Since March 3 at 7:00 AM UTC, an issue has been affecting some calls using the Conversational v3 Model, which is in Alpha stage. We have identified the root cause, and a fix is being deployed. Conversation recordings, transcripts, and analytics are unavailable for affected calls. Affected calls are currently displayed as "in progress" and will be transitioned to "failed" automatically in the next 48h. We will provide a further update once the fix has been fully rolled out. Affected components Conversations (Degraded performance)

Status: Identified The issue has been identified as originating from the cloud provider. We are actively working with them to resolve it. Affected components Conversations (Degraded performance)

Status: Identified We have identified an underlying issue with our cloud provider that is impacting service availability. Affected components ElevenCreative (Partial outage)

Status: Identified We have isolated the components responsible for performance degradation and are attempting to implement workarounds to bypass the affected cloud provider services. Affected components UI (Partial outage) Speech to Text (Degraded performance) Other (Degraded performance) ElevenCreative (Partial outage) Other API endpoints (Degraded performance)

Status: Identified Our engineering team is continuing to investigate issues with the cloud provider components. Multiple engineering teams and leadership is involved in the efforts. We'll continue monitoring closely and provide timely updates as the situation evolves. Affected components UI (Partial outage) Speech to Text (Degraded performance) Other (Degraded performance) ElevenCreative (Partial outage) Other API endpoints (Degraded performance)

Status: Identified Our engineering team is actively exploring multiple remediation paths in parallel while working to isolate the root cause. We have escalated this issue both to our infrastructure provider and to internal leadership to ensure it receives the highest level of attention and resources. A definitive root cause has not yet been identified, but we are making iterative changes and closely monitoring their impact. We will continue to provide updates as our investigation progresses. Affected components Other (Degraded performance) ElevenCreative (Partial outage) Other API endpoints (Degraded performance) UI (Partial outage) Speech to Text (Degraded performance)

Status: Identified Our team continues to actively investigate the degraded performance affecting services. We are working through several lines of investigation in parallel and methodically testing potential contributing factors. This remains escalated with our infrastructure provider and internal leadership, with additional engineering resources dedicated to this effort. While a definitive root cause has not yet been identified, each step is helping us build a clearer picture of the underlying issue. We understand the impact this is having and are treating this with the highest priority. We will provide another update in 15-20 minutes. Affected components Other (Degraded performance) ElevenCreative (Partial outage) Other API endpoints (Degraded performance) UI (Partial outage) Speech to Text (Degraded performance)

Status: Identified We are continuing to investigate all possible root causes and eliminate potential causes. Additionally we are clarifying affected products below. What's affected: • Agents Platform - degraded performance due to issues around conversation initiation. Conversations that do start function as normal thus API customers should retry failed initiation conversations. • Website - intermittent slowness What's working normally: • Text-to-Speech (TTS) API • Speech-to-Text (STT) API • Dubbing API Our engineering team remains fully engaged and is continuing to work through active lines of investigation alongside our infrastructure provider. We are iterating on potential mitigations and monitoring the results closely. We will continue to provide updates every 15-20 minutes. Affected components Other (Degraded performance) ElevenCreative (Partial outage) Other API endpoints (Degraded performance) UI (Partial outage) Speech to Text (Degraded performance)

Status: Identified Our team remains fully engaged and continues to work through active lines of investigation. All previously noted affected and operational components remain in the same state. We will provide another update in 15-20 minutes Affected components Speech to Text (Operational) Other (Operational) ElevenCreative (Partial outage) Other API endpoints (Degraded performance) UI (Partial outage)

Status: Identified We detected that approximately 1.5% of v3 requests produced static noise Affected components Text to Speech (Degraded performance)

Status: Identified Degraded RAG performance, which may cause increased latency. Affected components RAG (Degraded performance)

Status: Identified At 18:30 UTC, an unintentional change resulted in conversation_id not being included as a field within Post Call Transcript webhooks. As a temporary workaround, we advise consuming it from conversation_initiation_client_data.dynamic_variables.system__conversation_id, if possible. We are currently working on a fix to address this. Affected components Other API endpoints (Partial outage)

Status: Identified Starting at ~14:00 UTC today, some conversations have been affected by a disruption in our data extraction and call summary mechanisms, and a subset of these may have seen increased conversation latency. We have identified the issue as stemming from reduced availability of the Gemini 2.0 and Gemini 2.5 Flash models, and are currently working to resolve the issue. Affected components Conversations (Degraded performance)

Status: Identified We have observed a significant reduction in errors, and continue to work on a solution to fully mitigate the core issue. Affected components Conversations (Degraded performance)

Status: Identified We have observed an increase in latency affecting the Gemini models and are actively investigating the issue, and are working on improvements to the formatting of the recently implemented Data Collection backups. Affected components Conversations (Degraded performance)

Status: Identified We have observed a reoccurrence of Data Collection failures, and are working on additional measures to mitigate them. Affected components Conversations (Degraded performance)