01 Analyse Text

You can use the SAP Conversational AI API to analyse your text or your audio file, and extract useful information from it.

You can jump at the end of this page if you're looking for more details on the Response returned by the calls to analyse_text and analyse_file methods.

Usage

Start by instantiating either a Client or a Request object, as shown below:

import sapcai

client = new sapcai.Client('YOUR_TOKEN')
request = client.request

# ...is the same as...

request = sapcai.Request('YOUR_TOKEN')

Text analysis

Use the analyse_text method of the client, and after a call to our API, we return a Response object containing the enriched entities extracted from your text, the intents detected and some additional information.

Please note that the length of the text is limited to 512 chars by call.

response = request.analyse_text('YOUR_TEXT')

# get the intent detected
intent = response.intent

# get all the location entities extracted from your text
locations = response.all('location')

# Do your code

[DEPRECATED] Audio analysis

Use the analyse_file method of the client, and after a call to our API, we return you a Response object containing the enriched entities extracted from your speech, the intents detected and some other useful information.

Please note that the length of the audio is limited to 10 seconds by call.

response = request.analyse_file('YOUR_FILE')

# get the intent detected
intent = response.intent

# get all the location entities extracted from your text
locations = response.all('location')

# Do your code

Response

A Response is generated after a call to either analyse_text or analyse_file Request methods.

Attributes

An instance of the Response class provides each of the following attributes:

Attributes	Type
raw	String: the raw unparsed json response
uuid	String: the uuid of the request
source	String: the user input
intents	Array[object]: all the matched intents
act	String: the act of the processed sentence
type	String: the type of the processed sentence
sentiment	String: the sentiment of the processed sentence
entities	Array[Entity]: the array of entities
language	String: the language of the input
processing_language	String: the language used to process the input
version	String: the version of the json
timestamp	String: the timestamp at the end of the processing
status	String: the status of the response

Methods

Get the first detected intent

Method	Params	Return
intent		Object: the first detected intent

response = request.analyse_text('YOUR_TEXT')

intent = response.intent

if (intent.slug === 'geetings' && intent.confidence > 0.7):
	# Do your code

Get one entity

Method	Params	Return
get(name)	name: String	Entity: the first Entity matched

response = request.analyse_text('YOUR_TEXT')

location = response.get('location')

Get all entities matching name

Method	Params	Return
all(name)	name: String	Array[Entity]: all the Entities matched

response = request.analyse_text('YOUR_TEXT')

locations = response.all('location')

Act helpers

Method	Params	Return
is_assert		Bool: whether or not the act is an assertion
is_command		Bool: whether or not the act is a command
is_wh_query		Bool: whether or not the act is a question
is_yn_query		Bool: whether or not the act is a query

Type helpers

Method	Params	Return
is_abbreviation		Bool: whether or not the sentence is asking for an abbreviation
is_entity		Bool: whether or not the sentence is asking for an entity
is_description		Bool: whether or not the sentence is asking for a description
is_Human		Bool: whether or not the sentence is asking for a human
is_location		Bool: whether or not the sentence is asking for a location
is_number		Bool: whether or not the sentence is asking for a number

Sentiment helpers

Method	Params	Return
is_vpositive		Bool: whether or not the sentiment is very positive
is_positive		Bool: whether or not the sentiment is positive
is_neutral		Bool: whether or not the sentiment is neutral
is_negative		Bool: whether or not the sentiment is negative
is_vnegative		Bool: whether or not the sentiment is very negative

More

For more information, do not hesitate to dive deeper in the code, as it is commented.