Configuring 3rd party integrations

Send utterances, transcripts, and call recordings to other applications

Send to Slack

[SendToSlack]
action = SlackLiveUtterances
formatter = TextFormatter
url = https://hooks.slack.com/services/xxx/xxx/xxx
username = xxx
showphonenumber = true
showaudiochannel = true

Table 1. `[SendToSlack]` parameters
Parameter	Values	Description
action	`SlackLiveUtterance`	The value specified here must be `SlackLiveUtterance` .
formatter	`TextFormatter` , `JsonFormatter`	For readable results, the value specified here must be `TextFormatter` .
url	URL: HTTP or HTTPS	Specify the Slack endpoint to use.
username	string	Specify the Slack username to use.
`channel`	string, `#dttdemo` (default)	Specify the Slack channel to use.
showphonenumber	`true` , `false`	Displays a phone number prefix in Slack for each utterance.
showaudiochannel	`>` , `<` , `?`	Show a call direction prefix in Slack: `>` for inbound, `<` for outbound, and `?` when unknown.

Send transcripts to V-Spark

[PostToSpark]
action = PostZipToVSpark
formatter = JsonFormatter
destination = http://localhost:3000/transcribe/xxx/xxx
authtoken = 
reloadflow = true
ext = .mp3
mono = true
bitrate = 24
retry = 2
keepTmp = false
tempdir = /dev/shm

Table 2. `[PostToVSpark] parameters`
Parameter	Values	Description
destination	URL: HTTP or HTTPS	Specify the V-Spark folder to use.
authtoken	string	The authentication token required for the `destination` folder.
reloadflow	`true` , `false`	`true` enables the transcript and audio to load on V-Spark without any changes. `false` enables the metadata and audio to load to V-Spark and transcribe as if it is a new call. Transcripts created before the data was received by V-Spark aren't used.
ext	`.wav` , `.mp3`	Specify the audio format to submit to V-Spark. The value specified here must be set to `.wav` if `reloadflow` is set to `false`. Formats other than `.wav` enable `reloadflow`.
mono	`true` , `false`	When `ext=.mp3` , sets the audio transcoding type to `mono` .
bitrate	integer	When `ext=.mp3` , sets the audio bit rate used for audio transcoding.
retry	integer	Specify the number of times to retry a failed transcoding.
keepTmp	`true` , `false`	If `true` , temporary data is kept after sending.
tempdir	/path/to/temp/storage/	Sets the root directory for temporary storage to /tempdir/vociadapter/.

Send call data to Medallia

[SendToMedallia]
action      = SendCallDataToMedallia
formatter   = TextFormatter
speechurl   = 
tokenurl    = 
client      = speech_demo
secret      = 
minioserver = 
minioaccess = 
miniosecret = 
miniobucket = 
directory   = vociadapter
locale      = en-us
engine      = Engine 1 
apply_diarization = No
agent_channel = 1

Table 3. [SendToMedallia] parameters
Parameter	Values	Description
formatter	`TextFormatter`, `JsonFormatter`	Required for the Voci adapter.
speechurl	URL: HTTPS	Specify the URL of the Speech API on Medallia Experience Cloud (MEC).
tokenurl	URL: HTTPS	Specify the MEC URL that provides the authorization token.
client	string	MEC client username
secret	string	MEC client password
minioserver	URL: HTTPS	Specify the MinIO server URL.
minioaccess	string	Specify the username to access the MinIO server.
miniosecret	string	Specify the password to access the MinIO server.
miniobucket	string	Specify the name of the MinIO bucket to use.
directory	`path/to/audio/samples`	Specify the directory path in the MinIO bucket.
locale	string	Specify the language to use.
engine	`Engine 1`, `Engine 3`	Specify the engine to use. Medallia Speech supports two engines. `Engine 1` includes Voci-supported languages and models. `Engine 3` includes Speechmatics-supported languages.
apply_diarization	`No`, `Yes`	If set to `No`, the receiving system assumes two-channel audio. If set to `Yes`, the receiving system assumes mono audio and attempts to separate speakers.
agent_channel	`0`, `1`	Specify the audio channel the agent is speaking on.

Send call data with basic HTTP authentication

[SendCallDataWithBasicAuth]
action      = SendCallDataWithBasicAuth
transcribe  = true
username    = johndoe
password    = password
#basicauth   = johndoe:password
uri         = hostname.com/path/to/endpoint

Table 4. [SendCallDataWithBasicAuth] parameters
Parameter	Values	Description
transcribe	`true`, `false`	Enables V-Blaze transcription and includes transcripts with the call data.
username	string	Specify the basic HTTP authentication username. Do not specify a value for `basicauth` when using `username` and `password` to set access credentials.
password	string	Specify the basic HTTP authentication password. Do not specify a value for `basicauth` when using `username` and `password` to set access credentials.
basicauth	string	Use as an alternative to providing the `username` and `password` credentials. Do not specify a value for `basicauth` when using `username` and `password` to set access credentials.
uri	URL: HTTP or HTTPS	Specify the URL to send the call data.

Upload call data to S3

[UploadCallDataToS3]
action          = UploadCallDataToS3
transcribe      = true
audio           = true
audio_filename_override = false
cti_metadata    = true
bucket          = 
accesskey       = 
secret          =

Table 5. [UploadCallDataToS3] parameters
Parameter	Values	Description
transcribe	`true`, `false`	Enables V-Blaze transcription and includes transcripts with the call data.
audio	`true`, `false`	Includes the audio with the call data.
audio_filename_override	`true`, `false`	Overwrites the audio filename based on metadata.
cti_metadata	`true`, `false`	Includes all OrecX call metadata in the JSON transcript.
bucket	URL	Specify the URL of the s3 bucket. Note: Comment out the `accesskey` and `secret` parameters if the s3 bucket does not have security settings.
accesskey	string	Specify the access key of the s3 bucket. Comment out this parameter if the specified s3 bucket does not have an access key.
secret	string	Specify the secret of the s3 bucket. Comment out this option if the specified s3 bucket does not have a secret.

Send utterance data to XSell

[LivePostToXSell]
action    = SendUttDataToXSell
formatter = TextFormatter
clientid  = 
url       = 
apitoken  = 
livepost  = true
#Enable logging the utterance text
logtext   = true

Table 6. [LivePostToXSell] parameters
Parameter	Values	Description
formatter	`TextFormatter`, `JsonFormatter`	For readable results, the value specified here must be `TextFormatter` .
clientid	string	Specify the username.
url	URL: HTTP or HTTPS	URL endpoint to send call data.
apitoken	string	Specify the API authorization token.
logtext	`true`, `false`	Enables utterance logging to log file. Prints `--NOT LOGGED--` to the log file when set to `false`.
livepost	`true`, `false`	Enables to transcribes live-call audio streams and send the complete transcript when the call ends.

Note:

Contact your XSELL Technologies representative for clientid, url, and apitoken values.