For the complete documentation index, see llms.txt. Any page on this site can be fetched as Markdown by appending .md to its URL.

Voice agentsAgents

Get the speech-recognition tuning status for an agent's vocabulary

GET
/v1/agent-media/agents/{agentId}/vocabulary-status

Whether speech recognition has been tuned for the published version's vocabulary terms. Tuning starts when a version is published and usually takes a few minutes; asking also starts it if it has not started, so this is safe to poll. The agent uses its terms in its instructions whatever this reports.

state is READY (in use on calls), PENDING (being built), FAILED (see failureReason), SKIPPED_CAP (not available right now), NONE (no published terms that can be tuned), or DISABLED (not offered in this environment).

Authorization

bearerAuth
AuthorizationBearer <token>

A Tandem bearer token. Requests are scoped to the organization the token belongs to; no organization header is required.

In: header

Path Parameters

agentId*string

The agent's unique ID.

Response Body

application/json

application/json

application/json

application/json

curl -X GET "https://example.com/v1/agent-media/agents/string/vocabulary-status"
{  "status": {    "state": "NONE",    "failureReason": "string"  }}
{  "error": "string",  "message": "string",  "limit": 0,  "length": 0,  "reason": "string"}
{  "error": "string",  "message": "string",  "limit": 0,  "length": 0,  "reason": "string"}
{  "error": "string",  "message": "string",  "limit": 0,  "length": 0,  "reason": "string"}