Get reliability settings

Read the workspace's provider fallbacks for voice, LLM and speech-to-text.

GEThttps://api.tried.com.au/workspace/reliability

Returns the workspace's Reliability settings: which voice, LLM and STT provider to fall back to when the primary one fails. Any role can read them. See Reliability.

Request#

This endpoint takes no path, query or body parameters.

Example request#

curl https://api.tried.com.au/workspace/reliability \
  -H "Authorization: Bearer $TRIED_API_KEY"

const res = await fetch("https://api.tried.com.au/workspace/reliability", {
  headers: { Authorization: `Bearer ${process.env.TRIED_API_KEY}` },
});
const reliability = await res.json();

import os, requests

res = requests.get(
    "https://api.tried.com.au/workspace/reliability",
    headers={"Authorization": f"Bearer {os.environ['TRIED_API_KEY']}"},
)
reliability = res.json()

Response#

200 OK

{
  "fallback_voice_id": null,
  "llm_fallback_model": null,
  "stt_fallback_provider": null,
  "llm_fallback_timeout_ms": 4000
}

All fallbacks are null (off) until you set them.

Field	Type	Description
`fallback_voice_id`	string or null	Voice used when the agent's voice and its own `fallback_voice_ids` fail.
`llm_fallback_model`	string or null	Model used when the agent's LLM errors or times out.
`stt_fallback_provider`	string or null	STT provider used when the primary one fails: `deepgram`, `assemblyai`, `soniox`, `azure` or `openai`.
`llm_fallback_timeout_ms`	integer	Retry an LLM turn on the fallback model after this many ms without a first token. Default `4000`.

Errors#

Status	When
`401`	Missing or invalid API key.
`403`	The key's IP allowlist or scopes reject the request. This route needs the `admin` scope.
`429`	Rate limit exceeded. See Rate limits.

See Errors for the body format.