#I'm getting an error 429 when fetching transcripts
1 messages · Page 1 of 1 (latest)
A 429 from the ElevenLabs API means one of two things:
- too_many_concurrent_requests: you’ve exceeded your plan’s concurrency. Reduce parallel requests (including polling GETs) or upgrade. For Speech-to-Text, the current STT concurrency limits are: Free 8, Starter 12, Creator 20, Pro 40, Scale 60, Business 60. (help.elevenlabs.io)
- system_busy: the platform is at capacity. Retry with exponential backoff; higher tiers receive higher priority. (elevenlabs.io)
If you’re “fetching transcripts” via GET /v1/speech-to-text/transcripts/:transcription_id and seeing 429s, the most common causes are polling too aggressively or fetching many transcripts in parallel. Mitigations:
- Prefer webhooks for STT results to avoid polling entirely (set webhook=true when creating the transcription and handle the callback). (elevenlabs.io)
- If you must poll, cap parallel GETs below your plan’s concurrency and add jittered exponential backoff on 429. (elevenlabs.io)
- Check the status page if you suspect capacity issues (e.g., recent STT incidents have temporarily returned 429/500). (status.elevenlabs.io)
References:
- Error 429 overview and meanings. (help.elevenlabs.io)
- STT concurrency limits and capabilities. (elevenlabs.io)
- STT create and get transcript endpoints. (elevenlabs.io)
- Status page example for STT capacity. (status.elevenlabs.io)
If you can share the exact 429 message ("too_many_concurrent_requests" vs "system_busy") and your plan tier, I can suggest precise request rates.
Sources:
API - Error Code 429 – ElevenLabs
Error messages | ElevenLabs Documentation
Create transcript | ElevenLabs Documentation
Error messages | ElevenLabs Documentation
Limited capacity for Speech-To-Text - ElevenLabs Status
API - Error Code 429 – ElevenLabs
Speech to Text | ElevenLabs Documentation
Create transcript | ElevenLabs Documentation
Limited capacity for Speech-To-Text - ElevenLabs Status
i did upgrade to the business and its still doing it
Thanks for the update. A 429 on transcripts after upgrading to Business almost always means one of two things:
- too_many_concurrent_requests — you’re exceeding your plan’s concurrency
- system_busy — temporary platform load; retry with backoff
How to pinpoint and fix:
- Check the 429 message
- Inspect the response body to see whether it’s too_many_concurrent_requests or system_busy. (elevenlabs.io)
- If it’s too_many_concurrent_requests
- Verify your concurrency against plan limits:
- Speech to Text (Scribe): Business = 60 concurrent transcriptions. (elevenlabs.io)
- Other API workloads: Business = 15 concurrent requests (30 for Flash/Turbo models). (help.elevenlabs.io)
- Be aware that STT internally parallelizes long files (>8 minutes) by chunking into up to 4 segments processed concurrently; running many long files at once increases parallel load. Consider queueing jobs client‑side. (elevenlabs.io)
- Prefer webhooks (webhook=true) to receive results instead of polling, which reduces concurrent pressure on your side. (elevenlabs.io)
- If it’s system_busy
- Retry with exponential backoff and jitter. This is the recommended handling for 429 system_busy. (elevenlabs.io)
- Check the status page for any ongoing incidents that could elevate 429s. (status.elevenlabs.io)
- If you’re using the api-global-preview domain, note it has limited capacity and can increase 429s; try the standard API hostname. (elevenlabs.io)
- Still stuck?
- Please share one failing request’s x-trace-id and endpoint so we can escalate; we recommend logging x-trace-id for all requests. (elevenlabs.io)
If you can paste the exact 429 payload (message string), the endpoint you’re calling, how many requests you fire in parallel, and typical file durations, I can give plan‑specific guidance. (elevenlabs.io)
Sources:
Error messages | ElevenLabs Documentation
Speech to Text | ElevenLabs Documentation
How many requests can I make and can I increase it? – ElevenLabs
Speech to Text | ElevenLabs Documentation
Asynchronous Speech to Text | ElevenLabs Documentation
Error messages | ElevenLabs Documentation
Text-to-Voice Elevated Error Rate - ElevenLabs Status
Latency optimization | ElevenLabs Documentation
Latency optimization | ElevenLabs Documentation
Error messages | ElevenLabs Documentation
14|elevenl | [API] Request failed with status 429
14|elevenl | [API] Error response: <!doctype html><meta charset="utf-8"><meta name=viewport content="width=device-width, initial-scale=1"><title>429</title>429 Too Many Requests
14|elevenl | [API] Request URL: https://api.elevenlabs.io/v1/convai/conversations?page_size=50
14|elevenl | [API] API key being used: sk_f5332...
14|elevenl | [API] Error fetching conversations: Error: API Error: 429 - <!doctype html><meta charset="utf-8"><meta name=viewport content="width=device-width, initial-scale=1"><title>429</title>429 Too Many Requests
14|elevenl | at fetchConversations (file:///var/www/vhosts/igniteups.ai/apis/elevenlabs.js:88:13)
14|elevenl | at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
14|elevenl | at async syncConversations (file:///var/www/vhosts/igniteups.ai/apis/elevenlabs.js:573:20)
14|elevenl | [Sync] Rate limit exceeded after all retries. Stopping sync. Will retry at next interval.