Speech-to-text audio transcription is in limited early release and must be enabled by Vivox. For pricing information and to enable this service for your organization, contact your sales representative.
To enable speech-to-text transcription in a channel, issue a
vx_req_session_transcription_control
request with the enable field set to 1 and the
session_handle
field set to the handle of the already joined channel. The Vivox access token (the
for the local user, so it's the single sufficient condition. Calling the request before media is connected will return error 1028 (
VX_E_SESSION_DOES_NOT_HAVE_AUDIO
).
The Vivox SDK responds to the
vx_req_session_transcription_control
request with a
vx_resp_session_transcription_control
response. The return_code parameter is 0 on success, otherwise it returns a non-zero parameter.The following code is an example of how to enable speech-to-text transcription.
. . .vx_req_session_transcription_control_t *req;vx_req_session_transcription_control_create(&req);req->session_handle = vx_strdup("mychannel");req->enable = 1;req->access_token = vx_strdup (access_token.c_str());vx_issue_request2(&req->base);. . .void HandleResponse(vx_resp_session_transcription_control *resp){ if (resp->base.return_code != 0) { printf("Failed to set transcription status (status: %s) \n", vx_get_error_string(resp->base.status_code)); } else { printf("Transcription status successfully updated\n"); }}. . .