TranscribeService / Client / create_call_analytics_category
create_call_analytics_category#
- TranscribeService.Client.create_call_analytics_category(**kwargs)#
Creates a new Call Analytics category.
All categories are automatically applied to your Call Analytics transcriptions. Note that in order to apply categories to your transcriptions, you must create them before submitting your transcription request, as categories cannot be applied retroactively.
When creating a new category, you can use the
InputType
parameter to label the category as aPOST_CALL
or aREAL_TIME
category.POST_CALL
categories can only be applied to post-call transcriptions andREAL_TIME
categories can only be applied to real-time transcriptions. If you do not includeInputType
, your category is created as aPOST_CALL
category by default.Call Analytics categories are composed of rules. For each category, you must create between 1 and 20 rules. Rules can include these parameters: , , , and .
To update an existing category, see .
To learn more about Call Analytics categories, see Creating categories for post-call transcriptions and Creating categories for real-time transcriptions.
See also: AWS API Documentation
Request Syntax
response = client.create_call_analytics_category( CategoryName='string', Rules=[ { 'NonTalkTimeFilter': { 'Threshold': 123, 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'Negate': True|False }, 'InterruptionFilter': { 'Threshold': 123, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'Negate': True|False }, 'TranscriptFilter': { 'TranscriptFilterType': 'EXACT', 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'Negate': True|False, 'Targets': [ 'string', ] }, 'SentimentFilter': { 'Sentiments': [ 'POSITIVE'|'NEGATIVE'|'NEUTRAL'|'MIXED', ], 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'Negate': True|False } }, ], InputType='REAL_TIME'|'POST_CALL' )
- Parameters:
CategoryName (string) –
[REQUIRED]
A unique name, chosen by you, for your Call Analytics category. It’s helpful to use a detailed naming system that will make sense to you in the future. For example, it’s better to use
sentiment-positive-last30seconds
for a category over a generic name liketest-category
.Category names are case sensitive.
Rules (list) –
[REQUIRED]
Rules define a Call Analytics category. When creating a new category, you must create between 1 and 20 rules for that category. For each rule, you specify a filter you want applied to the attributes of a call. For example, you can choose a sentiment filter that detects if a customer’s sentiment was positive during the last 30 seconds of the call.
(dict) –
A rule is a set of criteria that you can specify to flag an attribute in your Call Analytics output. Rules define a Call Analytics category.
Rules can include these parameters: , , , and .
To learn more about Call Analytics rules and categories, see Creating categories for post-call transcriptions and Creating categories for real-time transcriptions.
To learn more about Call Analytics, see Analyzing call center audio with Call Analytics.
Note
This is a Tagged Union structure. Only one of the following top level keys can be set:
NonTalkTimeFilter
,InterruptionFilter
,TranscriptFilter
,SentimentFilter
.NonTalkTimeFilter (dict) –
Flag the presence or absence of periods of silence in your Call Analytics transcription output. Refer to for more detail.
Threshold (integer) –
Specify the duration, in milliseconds, of the period of silence that you want to flag. For example, you can flag a silent period that lasts 30,000 milliseconds.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for a period of silence. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for a period of silence. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
Negate (boolean) –
Set to
TRUE
to flag periods of speech. Set toFALSE
to flag periods of silence
InterruptionFilter (dict) –
Flag the presence or absence of interruptions in your Call Analytics transcription output. Refer to for more detail.
Threshold (integer) –
Specify the duration of the interruptions in milliseconds. For example, you can flag speech that contains more than 10,000 milliseconds of interruptions.
ParticipantRole (string) –
Specify the interrupter that you want to flag. Omitting this parameter is equivalent to specifying both participants.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for an interruption. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for an interruption. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
Negate (boolean) –
Set to
TRUE
to flag speech that does not contain interruptions. Set toFALSE
to flag speech that contains interruptions.
TranscriptFilter (dict) –
Flag the presence or absence of specific words or phrases in your Call Analytics transcription output. Refer to for more detail.
TranscriptFilterType (string) – [REQUIRED]
Flag the presence or absence of an exact match to the phrases that you specify. For example, if you specify the phrase “speak to a manager” as your
Targets
value, only that exact phrase is flagged.Note that semantic matching is not supported. For example, if your customer says “speak to the manager”, instead of “speak to a manager”, your content is not flagged.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for the specified key words or phrases. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for the specified key words or phrases. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
ParticipantRole (string) –
Specify the participant that you want to flag. Omitting this parameter is equivalent to specifying both participants.
Negate (boolean) –
Set to
TRUE
to flag the absence of the phrase that you specified in your request. Set toFALSE
to flag the presence of the phrase that you specified in your request.Targets (list) – [REQUIRED]
Specify the phrases that you want to flag.
(string) –
SentimentFilter (dict) –
Flag the presence or absence of specific sentiments in your Call Analytics transcription output. Refer to for more detail.
Sentiments (list) – [REQUIRED]
Specify the sentiments that you want to flag.
(string) –
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for the specified sentiments. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for the specified sentiments. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
ParticipantRole (string) –
Specify the participant that you want to flag. Omitting this parameter is equivalent to specifying both participants.
Negate (boolean) –
Set to
TRUE
to flag the sentiments that you didn’t include in your request. Set toFALSE
to flag the sentiments that you specified in your request.
InputType (string) –
Choose whether you want to create a real-time or a post-call category for your Call Analytics transcription.
Specifying
POST_CALL
assigns your category to post-call transcriptions; categories with this input type cannot be applied to streaming (real-time) transcriptions.Specifying
REAL_TIME
assigns your category to streaming transcriptions; categories with this input type cannot be applied to post-call transcriptions.If you do not include
InputType
, your category is created as a post-call category by default.
- Return type:
dict
- Returns:
Response Syntax
{ 'CategoryProperties': { 'CategoryName': 'string', 'Rules': [ { 'NonTalkTimeFilter': { 'Threshold': 123, 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'Negate': True|False }, 'InterruptionFilter': { 'Threshold': 123, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'Negate': True|False }, 'TranscriptFilter': { 'TranscriptFilterType': 'EXACT', 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'Negate': True|False, 'Targets': [ 'string', ] }, 'SentimentFilter': { 'Sentiments': [ 'POSITIVE'|'NEGATIVE'|'NEUTRAL'|'MIXED', ], 'AbsoluteTimeRange': { 'StartTime': 123, 'EndTime': 123, 'First': 123, 'Last': 123 }, 'RelativeTimeRange': { 'StartPercentage': 123, 'EndPercentage': 123, 'First': 123, 'Last': 123 }, 'ParticipantRole': 'AGENT'|'CUSTOMER', 'Negate': True|False } }, ], 'CreateTime': datetime(2015, 1, 1), 'LastUpdateTime': datetime(2015, 1, 1), 'InputType': 'REAL_TIME'|'POST_CALL' } }
Response Structure
(dict) –
CategoryProperties (dict) –
Provides you with the properties of your new category, including its associated rules.
CategoryName (string) –
The name of the Call Analytics category. Category names are case sensitive and must be unique within an Amazon Web Services account.
Rules (list) –
The rules used to define a Call Analytics category. Each category can have between 1 and 20 rules.
(dict) –
A rule is a set of criteria that you can specify to flag an attribute in your Call Analytics output. Rules define a Call Analytics category.
Rules can include these parameters: , , , and .
To learn more about Call Analytics rules and categories, see Creating categories for post-call transcriptions and Creating categories for real-time transcriptions.
To learn more about Call Analytics, see Analyzing call center audio with Call Analytics.
Note
This is a Tagged Union structure. Only one of the following top level keys will be set:
NonTalkTimeFilter
,InterruptionFilter
,TranscriptFilter
,SentimentFilter
. If a client receives an unknown member it will setSDK_UNKNOWN_MEMBER
as the top level key, which maps to the name or tag of the unknown member. The structure ofSDK_UNKNOWN_MEMBER
is as follows:'SDK_UNKNOWN_MEMBER': {'name': 'UnknownMemberName'}
NonTalkTimeFilter (dict) –
Flag the presence or absence of periods of silence in your Call Analytics transcription output. Refer to for more detail.
Threshold (integer) –
Specify the duration, in milliseconds, of the period of silence that you want to flag. For example, you can flag a silent period that lasts 30,000 milliseconds.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for a period of silence. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for a period of silence. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
Negate (boolean) –
Set to
TRUE
to flag periods of speech. Set toFALSE
to flag periods of silence
InterruptionFilter (dict) –
Flag the presence or absence of interruptions in your Call Analytics transcription output. Refer to for more detail.
Threshold (integer) –
Specify the duration of the interruptions in milliseconds. For example, you can flag speech that contains more than 10,000 milliseconds of interruptions.
ParticipantRole (string) –
Specify the interrupter that you want to flag. Omitting this parameter is equivalent to specifying both participants.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for an interruption. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for an interruption. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
Negate (boolean) –
Set to
TRUE
to flag speech that does not contain interruptions. Set toFALSE
to flag speech that contains interruptions.
TranscriptFilter (dict) –
Flag the presence or absence of specific words or phrases in your Call Analytics transcription output. Refer to for more detail.
TranscriptFilterType (string) –
Flag the presence or absence of an exact match to the phrases that you specify. For example, if you specify the phrase “speak to a manager” as your
Targets
value, only that exact phrase is flagged.Note that semantic matching is not supported. For example, if your customer says “speak to the manager”, instead of “speak to a manager”, your content is not flagged.
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for the specified key words or phrases. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for the specified key words or phrases. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
ParticipantRole (string) –
Specify the participant that you want to flag. Omitting this parameter is equivalent to specifying both participants.
Negate (boolean) –
Set to
TRUE
to flag the absence of the phrase that you specified in your request. Set toFALSE
to flag the presence of the phrase that you specified in your request.Targets (list) –
Specify the phrases that you want to flag.
(string) –
SentimentFilter (dict) –
Flag the presence or absence of specific sentiments in your Call Analytics transcription output. Refer to for more detail.
Sentiments (list) –
Specify the sentiments that you want to flag.
(string) –
AbsoluteTimeRange (dict) –
Makes it possible to specify a time range (in milliseconds) in your audio, during which you want to search for the specified sentiments. See for more detail.
StartTime (integer) –
The time, in milliseconds, when Amazon Transcribe starts searching for the specified criteria in your audio. If you include
StartTime
in your request, you must also includeEndTime
.EndTime (integer) –
The time, in milliseconds, when Amazon Transcribe stops searching for the specified criteria in your audio. If you include
EndTime
in your request, you must also includeStartTime
.First (integer) –
The time, in milliseconds, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in milliseconds, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
RelativeTimeRange (dict) –
Makes it possible to specify a time range (in percentage) in your media file, during which you want to search for the specified sentiments. See for more detail.
StartPercentage (integer) –
The time, in percentage, when Amazon Transcribe starts searching for the specified criteria in your media file. If you include
StartPercentage
in your request, you must also includeEndPercentage
.EndPercentage (integer) –
The time, in percentage, when Amazon Transcribe stops searching for the specified criteria in your media file. If you include
EndPercentage
in your request, you must also includeStartPercentage
.First (integer) –
The time, in percentage, from the start of your media file until the specified value. Amazon Transcribe searches for your specified criteria in this time segment.
Last (integer) –
The time, in percentage, from the specified value until the end of your media file. Amazon Transcribe searches for your specified criteria in this time segment.
ParticipantRole (string) –
Specify the participant that you want to flag. Omitting this parameter is equivalent to specifying both participants.
Negate (boolean) –
Set to
TRUE
to flag the sentiments that you didn’t include in your request. Set toFALSE
to flag the sentiments that you specified in your request.
CreateTime (datetime) –
The date and time the specified Call Analytics category was created.
Timestamps are in the format
YYYY-MM-DD'T'HH:MM:SS.SSSSSS-UTC
. For example,2022-05-04T12:32:58.761000-07:00
represents 12:32 PM UTC-7 on May 4, 2022.LastUpdateTime (datetime) –
The date and time the specified Call Analytics category was last updated.
Timestamps are in the format
YYYY-MM-DD'T'HH:MM:SS.SSSSSS-UTC
. For example,2022-05-05T12:45:32.691000-07:00
represents 12:45 PM UTC-7 on May 5, 2022.InputType (string) –
The input type associated with the specified category.
POST_CALL
refers to a category that is applied to batch transcriptions;REAL_TIME
refers to a category that is applied to streaming transcriptions.
Exceptions