Aura – ai

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Aura Bot generic-ai-dialog-v1

Description of Aura bot generic-ai-dialog-v1, that allows managing several experiences for TV channels using AI services and complex logic resolution.

Introduction

The generic-ai-dialog-v1 allows OBs to implement search and recommendation use cases for TV channels based on AI services and complex logic resolution for channelData v1 channels.

You can check its source code in Github: generic-ai-dialog-v1.ts.

Environment variables

Property	Type	Description	Mandatory
AURA_AUTHORIZATION_HEADER	string	Authorization header to be sent to `AURA_GATEWAY_API_ENDPOINT`	yes
AURA_COMPLEX_LOGIC_ENDPOINT	string	URL of Complex Logic endpoint to interact with resolution API.	yes
AURA_GATEWAY_API_ENDPOINT	string	URL of Aura Gateway API endpoint	yes
AURA_GATEWAY_API_ISSUER_URL	string	Issuer URL for token info	yes
AURA_CHANNELDATA_DEFAULT_VERSION	string	Default version of channelData. By default `1.0.0`.	no
AURA_CHANNELDATA_LA_DEFAULT_VERSION	string	Default version of channelData for Latin America. By default `2.0.0`.	no
GENERIC_MP_RESOLUTION_NO_RESULT_CODE	string[]	List of resolution codes that indicate no results found. By default `['211', '462']`.	no

How it works

The dialog works in a similar way to the generic-ai-dialog-v3 but adapted to the structure of channelData v1.

Complex Logic proxy is used to interact with Complex Logic resolution API, which returns a response similar to Movistar Plus Resolution API. This adaptation is only valid for TVSearch responses for STB channel, because only a subset of fields has been mapped from normalized TV API to legacy TV API.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Aura Bot generic-ai-dialog-v3

Description of Aura bot generic-ai-dialog-v3, that allows managing several experiences for TV channels using AI services and complex logic resolution.

Introduction

The generic-ai-dialog-v3 allows OBs to implement search and recommendation use cases for TV channels based on AI services and complex logic resolution for channelData v3 channels.

You can check its source code in Github: generic-ai-dialog-v3.ts.

Environment variables

Property	Type	Description	Mandatory
AURA_AUTHORIZATION_HEADER	string	Authorization header to be sent to `AURA_GATEWAY_API_ENDPOINT`	yes
AURA_COMPLEX_LOGIC_ENDPOINT	string	URL of Complex Logic endpoint to interact with resolution API.	yes
AURA_GATEWAY_API_ENDPOINT	string	URL of Aura Gateway API endpoint	yes
AURA_GATEWAY_API_ISSUER_URL	string	Issuer URL for token info	yes
AURA_CHANNELDATA_DEFAULT_VERSION	string	Default version of channelData. By default `1.0.0`.	no
AURA_CHANNELDATA_LA_DEFAULT_VERSION	string	Default version of channelData for Latin America. By default `2.0.0`.	no

Adding configurations for the AI Service

To use the AI Service, you need to add the intent configurations to the channel settings. The configurations should be added in the following format:

{
    "atria": {
        "dialogs": {
            "intent.tv.retrieve": {
                "applicationId": "816bdab6-3ea3-4a77-bdea-12945d6d7053",
                "presetId": "5d06f26c-a8d8-4f45-a999-102aaa8b191b"
            }
        }
    }
}

With this configuration, when the generic-ia-dialog-v3 is called with the intent intent.tv.retrieve, it will use the specified applicationId and presetId to call the generative API.

How it works

Steps

The dialog starts by retrieving the intent information from the context and logging the start of the dialog.
Then, it calls the AI service to get a generative response based on the intent and the current session ID. The session ID is updated in the context for future requests.
Next, it calls the complex logic service to get a resolution based on the intent and entities from the generative response.
If the resolution indicates no content found, it redirects to the conversational-search dialog.
Otherwise, it builds an activity to be sent to the user using the intent result, intent settings, and resolution.
If any error occurs during these steps, it logs the error, sets the error in the context, and builds an activity from the error.
Finally, it sends the activity to the user and ends the dialog.

Sequence diagram

sequenceDiagram
    autonumber
    participant User
    participant Dialog
    participant AI Services
    participant Complex Logic
    participant Conversational Search
    Dialog->>Dialog: Retrieve intent information from context
    Dialog->>AI Services: Call AI service with intent and session ID
    AI Services->>Dialog: Return generative response and session ID
    Dialog->>Complex Logic: Call complex logic service with intent and entities
    Complex Logic->>Dialog: Return resolution
    alt No content found in resolution
        Dialog->>Conversational Search: Redirect to conversational-search dialog
    else Content found
        Dialog->>Dialog: Build activity with intent result, intent settings, and resolution
        Dialog->>User: Send activity to user
    end
    Dialog->>Dialog: End dialog

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV LLM Conversational None UC

Global Use Case developed by Aura Platform Team that informs users of the platform’s available actions, ensuring correct invocation for proper system understanding.

Introduction

The video LLM Conversational None use case is a global experience designed and developed by Aura Global Team that provides Telefónica customers guidance on gaining knowledge of available platform actions and how to perform them correctly.

Find additional information in the following documents:

Discover the Conversational None specifications and the particularities of the use case development.
Learn how to configure the conversational-none-dialog for Conversational None use case.
Access the TV Retrieve plugin for LLM None UC.
Discover the resources used by the Conversational None use case.
Find the AI configuration for each OB for the TV LLM Conversational None UC.

Specifications

Request-response model

The TV LLM Conversational None UC is adapted to both request-response V1 and to its normalized model V3.

Available channels

The TV LLM Conversational None UC can be enabled in any channel in both v1 and v3 and TV-related payloads.

Use case development

The TV LLM Conversational None UC includes the following components:

Understanding features

Intent: intent.conversational.none

Use case logic

Once Aura recognizes the user’s utterance, the UC is resolved via:

Aura Bot dialog

TV LLM Conversational None UC is built over conversational-none-dialog.
See the Dialog configuration for the required settings and bypass configuration.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV LLM Conversational Search UC

Global use case developed by Aura Platform Team that allows users searching for recommendations by holding a conversation with Aura

Introduction

The video LLM Conversational Search use case is a global experience designed and developed by Aura Global Team that allows Telefónica customers to hold a conversation with Aura for having TV content recommendations using a voice interface.

Find additional information in the following documents:

Discover the Conversational search specifications and the particularities of the use case development.
Learn how to configure the conversational-search-dialog for Conversational Search use case.
Access the TV Retrieve plugin for LLM Search UC.
Discover the resources used by the Conversational search use case.
Find the AI configuration for each OB for the TV LLM Conversational Search UC.

Specifications

Kernel API

To resolve the user’s request, Aura uses the normalized Video Contents Kernel API.

Request-response model

The Conversational search use case is adapted to both request-response normalized model v3 and v1.

Available channels

The Conversational search use case can be enabled in any channel in both v1 and v3 and TV-related payloads.

Use case development

The Conversational search use case includes the following components:

Understanding features

Intent: intent.conversational.search

Use case logic

Once Aura recognizes the user’s utterance, the UC is resolved via:

Aura Bot dialog

The Conversational search use case is built over conversational-search-dialog.
See the Dialog configuration for the required settings and bypass configuration.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Global use case developed by Aura Platform Team that allows users to get personalized TV/video recommendations in Aura

Introduction

The video LLM Recommend use case is a global experience designed and developed by Aura Global Team that allows Telefónica customers to ask Aura for content recommendations (live channels and catalog items) using a voice interface.

Find additional information in the following documents:

Discover the TV LLM Recommend specifications and the particularities of the use case development.
Learn how to configure the generic-dialog for TV LLM Recommend use case.
Access the TV LLM Recommend plugin.
Discover the resources used by the TV LLM Recommend use case.
Find the AI configuration for each OB for the TV Recommend use case.

Specifications

Kernel API

To resolve the user’s request, Aura uses the normalized Video Contents Kernel API.

Request-response model

The TV LLM Recommend UC is available only when using the normalized request-response model v3.

Available channels

The TV LLM Recommend UC can be enabled in any channel implementing the normalized v3 request-response model and TV-related payloads.

The normalized TV LLM Recommend UC includes personalized recommendations:

Suggest a single live channel (with a zap action).
Suggest a single catalog item (e.g., VOD).
Suggest a list of items when multiple results are available.
Return no content when the API returns no results.

Current limitations

Currently, the TV LLM Recommend UC does not include:

Fallback/contingency search (as in TV search).
Filters customized (as persons, genres, participants, …).
Third-party catalogs outside Aura/Kernel scope.

Use case development

The TV LLM Recommend UC includes these components:

Understanding features

Intent: intent.tv.recommend

Use case logic

Once Aura recognizes the user’s utterance, the UC is resolved via:

Aura Bot dialog

TV LLM Recommend UC is built over generic-dialog-v3.
See the Dialog configuration for the required settings.

Response

The response follows the standard video plugin response schema.

Where:

intent: input intent.
entities: input entities as received.
result_intent: usually MEDIA.SEARCH; CHANNEL.PLAY is used when returning a single live channel.
resources: List of response resources that includes three main parameters:
- type: Type of the response resource.
- name: Name of the response resource. Example: tv:video.recommend.multiple-results, tv:video.recommend.by-default.simple-result, tv:video.recommend.display.single-result, tv:video.recommend.no-results.
- params: This field returns any required parameter in the resource. Example: channelId or deviceId for tv:video.recommend.display.single-result.
payload: Information provided by the Kernel API, when we make the content get info request. If we receive a response, this field includes the following parameters:
- type: The value of this field depends on the type of data included in the field data (info returned by the API):
  - If it is a value: details
  - If it is a list: content_list
- data: It returns the information provided by the Kernel API.
status: Final status request. This field includes the following parameters:
- code: Status code.
- message: Status message, which describes the status code.
- params: Parameter that sends details of status. This field does not appear if it is empty.
actions: Actions to follow with the result of request. This plugin could include the following actions:
- REMOTEON.ZAPTO: Used to tune a live channel on STB.
conditions: conditions for actions to be applied.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV LLM Search UC

Global use case developed by Aura Platform Team that allows users to search specific TV content in Aura

Introduction

The video LLM Search use case is a global experience designed and developed by Aura Global Team that allows Telefónica customers to ask Aura for TV content (live content and catalog items) using a voice interface.

Find additional information in the following documents:

Discover the TV LLM Search specifications and the particularities of the use case development.
Learn how to configure the generic-dialog for TV LLM Search use case.
Access the TV Retrieve plugin for LLM Search UC.
Discover the resources used by the TV LLM Search use case.
Find the AI configuration for each OB for the TV LLM Search use case.

Specifications

Kernel API

To resolve the user’s request, Aura uses the normalized Video Contents Kernel API.

Request-response model

The TV LLM Search UC is adapted to the normalized request-response model v3.

Available channels

The TV LLM Search UC can be enabled in any channel implementing the normalized v3 request-response model and TV-related payloads.

TV LLM Search features

The normalized TV LLM Search UC includes basic retrieval features for entertainment content:

Search by channel (including live channels and third-party streaming platforms).
Search by person (e.g., presenters/actors).
Search by free-form query strings. (e.g., TV content titles, genres, subgenres, …).
Support for multiple values per entity.

When a single live content is unambiguously found, the plugin can trigger a channel tune action (REMOTEON.ZAPTO) to start playback on the STB.

Current limitations

Currently, the TV LLM Search UC does not include:

Contingency/fallback search mode (no secondary search if zero results).
Temporal filters (time intervals).
Third-party catalogs outside Aura/Kernel scope.

Use case development

The TV LLM Search UC includes the following components:

Understanding features

Intent: intent.tv.retrieve
Entities consumed by the plugin:

Entity	Example (utterance)	Notes
`ent.audiovisual_channel`	“Pon #0” / “Abrir Movistar LaLiga”	Multiple channels allowed via `canon` with `
`ent.audiovisual_person`	“Lo de Broncano”	Multiple persons allowed via `canon` with `
`ent.audiovisual_query`	“noticias Oviedo”	Multiple tokens allowed via `canon` with `

⚠️ NLP should provide the canon field concatenated with || (double pipe) when multiple values are recognized (see NLP configuration) ⚠️ Other entities (e.g., time, devices, etc.) are not used by this plugin and should be omitted for this UC.

Use case logic

Once Aura recognizes the user’s utterance, the UC is resolved via:

Aura Bot dialog

TV LLM Search UC is built over generic-dialog-v3.
See the Dialog configuration for the required settings.

Response

The response follows the standard video plugin response schema.

Where:

intent: input intent.
entities: input entities as received.
result_intent: usually MEDIA.SEARCH; CHANNEL.PLAY is used when returning a single live channel.
resources: List of response resources that includes three main parameters:
- type: Type of the response resource.
- name: Name of the response resource. Example: tv:video.retrieve.multiple-results, tv:video.retrieve.by-default.simple-result, tv:video.retrieve.display.single-result, tv:video.retrieve.no-results.
- params: This field returns any required parameter in the resource. Example: queryDisplayed.
payload: Information provided by the Kernel API, when we make the content get info request. If we receive a response, this field includes the following parameters:
- type: The value of this field depends on the type of data included in the field data (info returned by the API):
  - If it is a value: details
  - If it is a list: content_list
- data: It returns the information provided by the Kernel API.
status: Final status request. This field includes the following parameters:
- code: Status code.
- message: Status message, which describes the status code.
- params: Parameter that sends details of status. This field does not appear if it is empty.
actions: Actions to follow with the result of request. This plugin could include the following actions:
- REMOTEON.ZAPTO: Used to tune a live channel on STB.
conditions: conditions for actions to be applied.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV LLM Triage UC

Global use case developed by Aura Platform Team that routes unanswered or unclassified requests to different LLMs

Introduction

The video LLM Triage use case is a global experience designed and developed by Aura Global Team that routes unanswered or unclassified requests to different LLMs to improve the chances of returning relevant results to the user.

Find additional information in the following documents:

Find the AI configuration for each OB for the TV LLM Triage use case.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Description of Aura bot conversational-search-dialog, that is specifically designed to resolve the advanced TV recommendation using LLMs in ATRIA and complex logic resolution.

Introduction

The conversational-search-dialog is specifically designed to resolve the advanced recommendation, using LLMs in ATRIA. Specifically the LLM configured for TV Conversational Search UC.

You can check its source code in Github:

conversational-search-dialog.ts.
conversational-search-v3-dialog.ts. Currently, this dialog is available for Aura channelData V1 and V3 versions.

For this use case, the required specific configuration is set below:

{
            "channelDataVersion": "v3", // Only in Aura channelData V3 version
            "id": "conversational-search-dialog-v3",
            "allowAnonymous": true,
            "triggerConditions": [
                {
                    "intent": "intent.conversational.search",
                    "settings": {
                        "locales": {
                            "error": [
                                "core:flc.answer.error"
                            ]
                        },
                        "needTvResolution": true
                    }
                }
            ],
            "bypass": {
                "duration": 60,
                "payloadName": "openai",
                "initialData": {},
                "recognizersEnabled": true,
                "recognizersBreakIntents": {
                    "intent.tv.display": [],
                    "intent.navigation.section_show": []
                }
            }
        }

Specific values

Key	Value	Description
bypass.duration	60 (minutes)	This value overwrites the default aura-bot value for bypass lasting 10 minutes.
bypass.payloadName	openai	Name of the interface that will be used as bypass payload for this dialog.
bypass.initialData	empty	It should contain any data needed by the dialog to be initialized.
bypass.recognizersEnabled	true	This value overwrites the default aura-bot way of working with recognizers when it is in bypass mode. If it is set to `true`, recognizers are executed, and their result is stored in the bypass payload to be used, if needed, afterwards.
bypass.recognizersBreakIntents	Array of intents	Used to handle when the user requested any other use case during the execution of the TV Conversational Search UC.

Configuration by channel

For this dialog it is necessary to add the configuration of applicationId and presetId per channel and intent in the channel model.

For this use case, the required specific configuration per channel is set below:

{   
    "atria": {
        "dialogs": {
            "intent.conversational.search": {
                "applicationId": "816bdab6-3ea3-4a77-bdea-12945d6d7053",
                "presetId": "acdef02f-f810-4474-8143-6b2a04a042f8"
            }
        }
    }
}

Environment variables

Property	Type	Description	Mandatory
AURA_AUTHORIZATION_HEADER	string	Authorization header to be sent to `AURA_GATEWAY_API_ENDPOINT`	yes
AURA_COMPLEX_LOGIC_ENDPOINT	string	URL of Complex Logic endpoint to interact with resolution API.	yes
AURA_GATEWAY_API_ENDPOINT	string	URL of Aura Gateway API endpoint	yes
AURA_GATEWAY_API_ISSUER_URL	string	Issuer URL for token info	yes
AURA_CHANNELDATA_DEFAULT_VERSION	string	Default version of channelData. By default `1.0.0`.	no
AURA_CHANNELDATA_LA_DEFAULT_VERSION	string	Default version of channelData for Latin America. By default `2.0.0`.	no
CONVERSATIONAL_SEARCH_MAX_TAGS	number	Default number of max tags to manage in the dialog. By default `4`.	no
CONVERSATIONAL_SEARCH_TAGS_SCORE_THRESHOLD	number	Default number with minimum score threshold required for a tag to be considered relevant in a TV Conversational Search. Value of this variable oscillates between 0 and 1. By default `0`.	no

How it works

Steps

The dialog starts by init bypass mode.
Then, it checks if the iteration comes from the Triage recognizer, by setting the user’s text to the fixed phase of resource: tv-conversational-search:command.recommend and then calling the AI service.
If the iteration originated because there were no results, it will return the text obtained from the resource: tv-conversational-search:command.no-results to the user.
Then, it calls the AI service to get a generative response based on the intent and the current session ID. The session ID is updated in the context for future requests.
Next, with the response obtained from the LLM, the user will be composed of the following:
- If the generative response returns command info, It will be checked if the command is EXIT so the bypass will be closed and the resource: tv-conversational-search:command.exit will be returned to the user.
- If the generative response returns tags info, it calls the complex logic service to get a resolution based on the intent and tags from the generative response. If the tags are not valid or not meet a threshold or there are no results, the resource will be returned: tv-conversational-search:command.no-results. Instead, ff content exists, it will be returned in the channel data, but the text of the result shown to the user will always be the one obtained from generative response.
- If the generative response returns text info, this text is included in response of user, unless there are no results, as we indicated in the previous paragraph, a specific resource will be returned.
Finally, it sends the activity to the user and ends the dialog.
The bypass will only be closed if these two conditions are met:
- The ‘EXIT’ command is detected.
- A new user intent configured that match those configured in the dialog in field: recognizersBreakIntents.

Sequence diagram

sequenceDiagram
    autonumber
    participant User
    participant Dialog
    participant AI Services
    participant Complex Logic

    Dialog->>Dialog: Open dialog 
    Dialog->>Dialog: Init bypass mode and check interaction origin
    alt interaction origin is no-results
        Dialog->>User: Send activity to user with command no-results text
        Dialog->>Dialog: End dialog
    else interaction origin is triage
        Dialog->>Dialog: set text to send LLM with command recommend.
    end
    Dialog->>Dialog: continue bypass
    alt user intent recognized match with break intent configured
        Dialog->>Dialog: Close bypass
        Dialog->>User: Send activity to user with command exit text
    else user intent is not break intent configured
        Dialog->>Dialog: continue bypass
        Dialog->>AI Services: Call AI service with intent and session ID
        AI Services->>Dialog: Return generative response and session ID
        alt generative response contains a 'EXIT' command
            Dialog->>Dialog: Close bypass
            Dialog->>User: Send activity to user with command exit text
            Dialog->>Dialog: End dialog
        else generative response contains tags
            alt tags are valid
                Dialog->>Complex Logic: Call complex logic service with intent and tags
                Complex Logic->>Dialog: Return resolution
                alt No content found in resolution
                    Dialog->>User: Send activity to user with command no-results text
                else Content found
                    Dialog->>User: Send activity to user with text of generative response and resolution
                end
            else tags are invalid
                Dialog->>User: Send activity to user with command no-results text
            end
        else generative response contains text
                Dialog->>User: Send activity to user with text of generative response
        end
    end
    Dialog->>Dialog: End dialog

Docs: TV LLM Conversational None: Dialog Configuration

Mon, 01 Jan 0001 00:00:00 +0000

Configuration of the generic-ai-dialog for TV LLM Conversational None UC

How the generic-ai-dialog must be configured for this use case

Specific configuration

The TV LLM Conversational None UC is triggered by the generic-ai-dialog-v1 for channelData v1 or generic-ai-dialog-v3 for channelData v3. For this use case, the required specific configuration is set below:

Example v1:

    {
    "name": "generic-ai",
    "dialogs": [
        {
            "id": "generic-ai-dialog-v1",
            "onlyIn": [
                "set-top-box"
            ],
            "triggerConditions": [
                {
                    "intent": "intent.conversational.none",
                    "settings": {
                        "locales": {
                            "error": [
                                "tv.error.fpaOrCognitive"
                            ]
                        }
                    }
                }
            ]
        }
    ]
    }

Example v3:

    {
        "channelDataVersion": "v3",
        "id": "generic-ai-dialog-v3",
        "onlyIn": [
            "set-top-box",
            "global-video-platform"
        ],
        "triggerConditions": [
            {
                "intent": "intent.conversational.none",
                "settings": {
                    "locales": {
                        "error": [
                            "core:flc.answer.error"
                        ]
                    }
                }
            }
        ]
    }

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Configuration of the generic-ai-dialog for TV LLM Search UC

How the generic-ai-dialog must be configured for this use case

Specific configuration

The TV LLM Search UC is triggered by the generic-ai-dialog-v1 for channelData v1 or generic-ai-dialog-v3 for channelData v3. For this use case, the required specific configuration is set below:

Example v1:

    {
        "id": "generic-ai-dialog-v3",
        "onlyIn": [
            "movistar-plus"
        ],
        "triggerConditions": [
            {
                "intent": "intent.tv.retrieve",
                "settings": {
                    "locales": {
                        "error": [
                            "tv:tv.error.fpaOrCognitive"
                        ]
                    },
                    "payloadType": "tv"
                }
            }
        ]
    }

Example v3:

    {
        "id": "generic-ai-dialog-v1",
        "onlyIn": [
            "movistar-plus"
        ],
        "triggerConditions": [
            {
                "intent": "intent.tv.retrieve",
                "settings": {
                    "locales": {
                        "error": [
                            "tv:tv.error.fpaOrCognitive"
                        ]
                    },
                    "payloadType": "tv"
                }
            }
        ]
    }

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV Conversational Search plugin

Complex Logic Framework plugin for the TV Conversational Search use case

Introduction

The TV Conversational Search plugin resolves requests from a user to obtain personalized recommendations for audiovisual content from Telefónica catalogs.

Intent: intent.conversational.search
Main result intents: MEDIA.SEARCH (default), CHANNEL.PLAY (when zapping to a live channel).

Validate input request

This step validates that the input request fulfills all the requirements from the input schema.

If the input data is invalid, the resource returned is:

tv:video.model-validation.request.error.

Entities used for search

The plugin works with the following entity types:

Entity type	Usage in search	Notes
`ent.audiovisual_query`	Yes	Free text query, multiple values accepted with `\|\|`.

Multiple entities & values:
You can repeat the same entity type multiple times and/or provide several values in a single entity using the \|\| separator.

Building the search query

The plugin constructs two query strings:

search_query → technical query sent to Kernel API:
- Group query and channel entities in parentheses.
- Joins different entity types with AND.
- Query: split on ||, trim each value, and join with OR inside the parentheses (no quotes).
- Example:
```
(mystery OR series)
```
search_query_response → a simplified, user-readable query string included in the payload and telemetry.

If no entities are present, the plugin returns a no results response.

API video query (Handler 4P)

Plugin calls Handler4P.tv_4p_handler.get_user_content_search(...) with the following parameters:

user_id: data[aura_user][user_id]
administrative_number: data[app_context][user][account_number]
access_token: data[aura_user][access_token]
scopes: data[aura_user][scopes]
purposes: data[aura_user][purposes]
device_type: data[app_context][device][type]
catalog_types:
- Default: VOD, LIVE, L7D
- Allowed: VOD, LIVE, L7D, LCH, LSR
- Filtered according to allowed_catalog_types of the device
search_query: built as explained above
show_series: series
profile: data[app_context][user][video_profile_name] (optional; only if present to avoid API 501 errors)
commercialization_types: [SVOD]
max_quality: data[app_context][device][max_quality] (optional)
current_region: data[app_context][location][currentRegion] (optional)
order: personal

Response logic

Depending on the number and type of results, the plugin builds different responses:

1. No results

Plays negative sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.conversational.search.no-results
Status: SUCCESS.NO_CONTENT

2. One result

Live channel:
- Payload: { type: details, data: <channel info> }
- Action: REMOTEON.ZAPTO (with params channelId and deviceId)
- Plays positive sound
- result_intent: CHANNEL.PLAY
- Resource: tv:video.conversational.search.single-result
Non-live content (VOD, L7D, etc.):
- Payload: details of the single item
- Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
- result_intent: MEDIA.SEARCH
- Resource: tv:video.conversational.search.by-default.simple-result

3. Multiple results

Payload: list of items
Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.conversational.search.multiple-results (with param query_displayed)

Resources returned

The TV Recommend plugin may return the following resources:

tv:video.model-validation.request.error → input validation error
tv:video.api.answer.error → API Video error
tv:video.conversational.search.no-results → no content found
tv:video.conversational.search.by-default.simple-result → one non-live result
tv:video.conversational.search.multiple-results → multiple results
tv:video.conversational.search.display.single-result → one live channel (tune channel)

Payload and actions

Payload:
- type: details (single item) or content_list (multiple items)
- data: item(s) from Kernel API
- For multiple results: includes search_query (human-readable string)
Actions:
- Sound feedback (positive/negative)
- For a single live channel: {"name": "REMOTEON.ZAPTO", "params": {"channelId": "$channelId", "deviceId": "$deviceId"}, "target": "stb"}

Status codes

status: Final status request. This field includes the following parameters:
- code: Status code with value: SUCCESS, SUCCESS.NO_CONTENT or ERROR.INTERNAL.
- message: Status message, which describes the status code.
- params: Parameter that sends details of status. This field does not appear if it is empty.

Example behavior

“Recommend me something”
- for a single live channel result → CHANNEL.PLAY tune channel with live channel.
- for a multiple results → MEDIA.SEARCH with tv:video.conversational.search.multiple-results.
- for any results → tv:video.conversational.search.no-results.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Complex Logic Framework plugin for the TV Recommend use case

Introduction

The TV Recommend plugin resolves requests from a user to obtain personalized recommendations for audiovisual content from Telefónica catalogs.

Intent: intent.tv.recommend
Main result intents: MEDIA.SEARCH (default), CHANNEL.PLAY (when zapping to a live channel).

Validate input request

This step validates that the input request fulfills all the requirements from the input schema.

If the input data is invalid, the resource returned is:

tv:video.model-validation.request.error.

Entities used for search

The plugin works with the following entity types:

Entity type	Usage in search	Notes
`ent.audiovisual_channel`	Yes	Multiple values accepted, separated by `\|\|`.
`ent.audiovisual_person`	Yes	For participants (actors, directors, presenters, etc.), multiple values accepted with `\|\|`.
`ent.audiovisual_query`	Yes	Free text query, multiple values accepted with `\|\|`.

Multiple entities & values:
You can repeat the same entity type multiple times and/or provide several values in a single entity using the \|\| separator.

Building the search query

The plugin constructs two query strings:

search_query → technical query sent to Kernel API:
- Group query and channel entities in parentheses.
- Joins different entity types with AND.
- Query: split on ||, trim each value, and join with OR inside the parentheses (no quotes).
- Channel: trim each value, wrap each value in quotes, and join with OR inside the parentheses.
- Person: trim each value, wrap each value in quotes, and join with AND (no group parentheses).
- Example:
```
("La 1" OR "La 2") AND "Brad Pitt" AND (mystery OR series)
```
search_query_response → a simplified, user-readable query string included in the payload and telemetry.

If no entities are present, the plugin returns a no results response.

API video query (Handler 4P)

Plugin calls Handler4P.tv_4p_handler.get_user_content_search(...) with the following parameters:

user_id: data[aura_user][user_id]
administrative_number: data[app_context][user][account_number]
access_token: data[aura_user][access_token]
scopes: data[aura_user][scopes]
purposes: data[aura_user][purposes]
device_type: data[app_context][device][type]
catalog_types:
- Default: VOD, LIVE, L7D
- Allowed: VOD, LIVE, L7D, LCH, LSR
- Filtered according to allowed_catalog_types of the device
search_query: built as explained above
show_series: series
profile: data[app_context][user][video_profile_name] (optional; only if present to avoid API 501 errors)
commercialization_types: [SVOD]
max_quality: data[app_context][device][max_quality] (optional)
current_region: data[app_context][location][currentRegion] (optional)
order: personal

Response logic

Depending on the number and type of results, the plugin builds different responses:

1. No results

Plays negative sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.recommend.no-results
Status: SUCCESS.NO_CONTENT

2. One result

Live channel:
- Payload: { type: details, data: <channel info> }
- Action: REMOTEON.ZAPTO (with params channelId and deviceId)
- Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
- result_intent: CHANNEL.PLAY
- Resource: tv:video.recommend.display.single-result
Non-live content (VOD, L7D, etc.):
- Payload: details of the single item
- Plays positive sound
- result_intent: MEDIA.SEARCH
- Resource: tv:video.recommend.by-default.simple-result

3. Multiple results

Payload: content list and search_query (search query response string)
Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.recommend.multiple-results (with param query_displayed)

Resources returned

The TV Recommend plugin may return the following resources:

tv:video.model-validation.request.error → input validation error
tv:video.api.answer.error → API Video error
tv:video.recommend.no-results → no content found
tv:video.recommend.by-default.simple-result → one non-live result
tv:video.recommend.multiple-results → multiple results
tv:video.recommend.display.single-result → one live channel (tune channel)

Payload and actions

Payload:
- type: details (single item) or content_list (multiple items)
- data: item(s) from Kernel API
- For multiple results: includes search_query (human-readable string)
Actions:
- Sound feedback (positive/negative)
- For a single live channel: {"name": "REMOTEON.ZAPTO", "params": {"channelId": "$channelId", "deviceId": "$deviceId"}, "target": "stb"}

Status codes

status: Final status request. This field includes the following parameters:
- code: Status code with value: SUCCESS, SUCCESS.NO_CONTENT or ERROR.INTERNAL.
- message: Status message, which describes the status code.
- params: Parameter that sends details of status. This field does not appear if it is empty.

Example behavior

“Recommend me something”
- for a single live channel result → CHANNEL.PLAY tune channel with live channel.
- for a multiple results → MEDIA.SEARCH with tv:video.recommend.multiple-results.
- for any results → tv:video.recommend.no-results.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV Retrieve plugin for LLM Search UC

Complex Logic Framework TV Retrieve plugin for the TV LLM Search UC

Introduction

The TV Retrieve plugin for LLM Search UC resolves requests from a user to search and retrieve audiovisual content from Telefónica catalogs.

Intent: intent.tv.retrieve
Main result intents: MEDIA.SEARCH (default), CHANNEL.PLAY (when zapping to a live channel).

Validate input request

This step validates that the input request fulfills all the requirements from the input schema.

If the input data is invalid, the resource returned is:

tv:video.model-validation.request.error.

Entities used for search

The plugin works with the following entity types:

Entity type	Usage in search	Notes
`ent.audiovisual_channel`	Yes	Multiple values accepted, separated by `\|\|`.
`ent.audiovisual_person`	Yes	For participants (actors, directors, presenters, etc.), multiple values accepted with `\|\|`.
`ent.audiovisual_query`	Yes	Free text query, multiple values accepted with `\|\|`.

Multiple entities & values:
You can repeat the same entity type multiple times and/or provide several values in a single entity using the \|\| separator.

Building the search query

The plugin constructs two query strings:

search_query → technical query sent to Kernel API:
- Group query and channel entities in parentheses.
- Joins different entity types with AND.
- Query: split on ||, trim each value, and join with OR inside the parentheses (no quotes).
- Channel: trim each value, wrap each value in quotes, and join with OR inside the parentheses.
- Person: trim each value, wrap each value in quotes, and join with AND (no group parentheses).
- Example:
```
("La 1" OR "La 2") AND "Brad Pitt" AND (mystery OR series)
```
search_query_response → a simplified, user-readable query string included in the payload and telemetry.

If no entities are present, the plugin returns a no results response.

API video query (Handler 4P)

Plugin calls Handler4P.tv_4p_handler.get_user_content_search(...) with the following parameters:

user_id: data[aura_user][user_id]
administrative_number: data[app_context][user][account_number]
access_token: data[aura_user][access_token]
scopes: data[aura_user][scopes]
purposes: data[aura_user][purposes]
device_type: data[app_context][device][type]
catalog_types:
- Default: VOD, LIVE, L7D
- Allowed: VOD, LIVE, L7D, LCH, LSR
- Filtered according to allowed_catalog_types of the device
search_query: built as explained above
show_series: series
profile: data[app_context][user][video_profile_name] (optional; only if present to avoid API 501 errors)
commercialization_types: [SVOD]
max_quality: data[app_context][device][max_quality] (optional)
current_region: data[app_context][location][currentRegion] (optional)

Response logic

Depending on the number and type of results, the plugin builds different responses:

1. No results

Plays negative sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.retrieve.no-results
Status: SUCCESS.NO_CONTENT

2. One result

Live channel:
- Payload: { type: details, data: <channel info> }
- Action: REMOTEON.ZAPTO (with params channelId and deviceId)
- Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
- result_intent: CHANNEL.PLAY
- Resource: tv:video.retrieve.display.single-result
Non-live content (VOD, L7D, etc.):
- Payload: details of the single item
- Plays positive sound
- result_intent: MEDIA.SEARCH
- Resource: tv:video.retrieve.by-default.simple-result

3. Multiple results

Payload: content list and search_query (search query response string)
Plays positive sound unless the channel is disabled in the sound_disabled_channels configuration
result_intent: MEDIA.SEARCH
Resource: tv:video.retrieve.multiple-results (with param query_displayed)

Resources returned

The TV Retrieve plugin for LLM Search UC may return the following resources:

tv:video.model-validation.request.error → input validation error
tv:video.api.answer.error → API Video error
tv:video.retrieve.no-results → no content found
tv:video.retrieve.by-default.simple-result → one non-live result
tv:video.retrieve.multiple-results → multiple results
tv:video.retrieve.display.single-result → one live channel (tune channel)

Payload and actions

Payload:
- type: details (single item) or content_list (multiple items)
- data: item(s) from Kernel API
- For multiple results: includes search_query (human-readable string)
Actions:
- Sound feedback (positive/negative)
- For a single live channel: {"name": "REMOTEON.ZAPTO", "params": {"channelId": "$channelId", "deviceId": "$deviceId"}, "target": "stb"}

Status codes

status: Final status request. This field includes the following parameters:
- code: Status code with value: SUCCESS, SUCCESS.NO_CONTENT or ERROR.INTERNAL.
- message: Status message, which describes the status code.
- params: Parameter that sends details of status. This field does not appear if it is empty.

Example behavior

“Put La 1” → detects ent.audiovisual_channel with value "La 1" → single live channel result → CHANNEL.PLAY tune channel with live channel.
“I want something with Brad Pitt” → detects ent.audiovisual_person with value "Brad Pitt" → multiple results → MEDIA.SEARCH with tv:video.retrieve.multiple-results.
“Mystery series” → detects audiovisual_query with value "Mystery series" → if no results → tv:video.retrieve.no-results.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

AI configuration by OB

The following information applies to both Spain and Brazil; there will not be separate blocks for each OB.

Introduction

LLMs UCs are configured and located within the aura-config-provision repository. See the TV LLM Conversational None Preset (notice that in the url, the field [ob] needs to be replaced by [es] or [br]) LLM use cases have been designed to provide a fallback mechanism for user requests that are not currently understood, i.e. return a ‘None’ intent, or when no content is found in the video platform related to it, offering a second chance for classification and response generation.

Intent

The Conversational None Use Case has different invocation points:

From triage: when the LLM fails to map the user’s input to any known intent often due to ambiguous, incomplete or broad queries (“Uno hacia abajo”, “Sintoniza el canal”, “Cambia cómo suena”).
From failures or KOs occurring in other supported use cases.

Once the user’s input is classified as intent.conversational.none, the request is routed to the Conversational None Agent, an LLM-based conversational module designed to handle unclassified or unsupported queries with the goal of guiding the user toward valid flows or supported actions within the platform by suggesting possible next steps or rephrasing options.

The conversation is considered complete once the LLM returns a response to the user. Subsequent user inputs will be reprocessed through the full recognition pipeline, including both deterministic and LLM-based intent classifiers.

LLM output format

In the output format, the LLM returns a structured JSON response based on predefined Use Cases configured within the field use_cases in the preset (in args). These Use Cases follow a structured format that includes three fields separated by a pipe (|):

NAME (mandatory): The name of the Use Case.
DESCRIPTION (optional): A short explanation or instruction. This field may be empty.
EXAMPLE (mandatory): A representative user utterance for that Use Case.

Sample extract from use case information:

NAME	DESCRIPTION	EXAMPLE
abrir una sección	Para abrir una sección es necesario mencionar el nombre de la sección que deseas consultar. Algunas son: Deportes, Cine o Series.	Abrir la sección Deportes
Desativar o som		Silencie a TV

Based on this information and according to matching criteria with the user’s request, the LLM constructs its output response by returning two fields:

“speak”: intended for voice output to be spoken by the system.
“text”: displayed on screen as visual feedback within the Video Platform interface.

The content of these fields varies depending on the type of match between user’s input and the predefined Use Cases:

Single Use Case Match:
- If the Use Case has a description:
  - “speak”: this field references the name of the Use Case related to the request, describes its invocation process and includes a relevant example.
  - “text”: this field mirrors the example used in “speak”.
- If the Use Case has no description:
  - “speak”: this field mentions the name of the Use Case linked to the request and the corresponding example.
  - “text”: this field mirrors the example used in “speak”.
Multiple Use Cases Match:
- “speak”: this field mentions the ambiguity and shows two likely Use Cases with examples.
- “text”: this field includes the same two examples.
No match:
- “speak”: this field explains the uncertainty and suggests two random valid Use Cases with examples.
- “text”: this field repeats only the examples pertaining to the Use Cases shown in “speak”.

This conditional behavior ensures that the assistant gives relevant, helpful responses tailored to the context of the user’s request.

Examples in Castilian Spanish:

Type of match	Input	Output
“Single Use Case no description”	“Pon la peli desde el principio”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Parece que quieres poner un contenido desde el principio. Por ejemplo, prueba a decir Pon Titanic desde el principio.", "text": "Pon Titanic desde el principio"}}}}`
“Single Use Case with description”	“Pon cualquier canal”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Parece que quieres cambiar de canal. Para cambiar de canal es necesario mencionar el nombre del canal que quieres sintonizar o indicar el número de dial asociado a ese canal. Por ejemplo, prueba a decir Pon el canal Antena 3.", "text": "Pon el canal Antena 3"}}}}`
“Multiple Use Cases”	“Volumen”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Creo que quieres ejecutar alguna acción relacionada con el volumen. Podría ser bajar el volumen o subir el volumen. Por ejemplo, prueba a decir Baja un poco el volumen o Sube un poco el volumen", "text": "Baja un poco el volumen, Sube un poco el volumen"}}}}`
“No match”	“Cántame una canción”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "No tengo claro a qué te refieres pero sí que puedo cambiar de canal o abrir una sección. Por ejemplo, prueba a decir Pon el canal Antena 3 o Abrir la sección Deportes", "text": "Pon el canal Antena 3, Abrir la sección Deportes"}}}}`

Examples in Brazilian Portuguese:

Type of match	Input	Output
“Single Use Case no description”	“Coloque o filme desde o começo”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Parece que você quer colocar um conteúdo desde o começo. Por exemplo, tente dizer Começar desde o início.", "text": "Começar desde o início"}}}}`
“Single Use Case with description”	“Coloque qualquer canal	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Parece que você quer mudar de canal. Para mudar de canal é necessário mencionar o nome do canal que deseja sintonizar. Por exemplo, tente dizer Mudar para o SBT.", "text": "Mudar para o SBT"}}}}`
“Multiple Use Cases”	“Volume”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Creo que você quer executar alguma ação relacionada ao volume. Pode ser abaixar o volume ou subir o volume. Por exemplo, tente dizer Abaixe o volume ou Suba o volume", "text": "Abaixe o volume, Suba o volume"}}}}`
“No match”	“Canta-me uma canção”	`{{"name": "llm_conversational_none", "arguments": {{ "speak": "Não tenho certeza do que você está se referindo mas eu posso mudar de canal ou abrir uma seção. Por exemplo, tente dizer Mudar para o SBT ou Abra Filmes", "text": "Mudar para o SBT, Abra Filmes"}}}}`

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

AI configuration by OB

The following information applies to both Spain and Brazil; there will not be separate blocks for each OB.

Introduction

LLMs UCs are configured and located within the aura-config-provision repository. See the TV LLM Conversational Search Preset (notice that in the url, the field [ob] needs to be replaced by [es] or [br]) LLM use cases have been designed to provide a fallback mechanism for user requests that are not currently understood, i.e. return a ‘None’ intent, or when no content is found in the video platform related to it, offering a second chance for classification and response generation.

Intent

The Conversational Search use case has different invocation points:

From triage: when the LLM identifies that the user’s intention is to perform an open search or to get an advanced recommendation (“Recomiéndame algo para ver”, “Quiero ver algo divertido”, “Ponme con el recomendador”)
Fallback for the TV Search deterministic use case and for the TV LLM Search UC for user’s requests with 0 results or KO.

Once the user’s request is classified under intent.conversational.search, the user will enter into a conversation with the LLM Conversational Search agent. During this conversation, the agent will attempt to refine the user’s search criteria to try to lead the user to content that is available on the platform according to the available thematic categories also called tags (e.g., “Amor imposible”, “Persecuciones”, “Psicópatas”).

The conversation can be ended through various means:

By breaking intents:

ES: intent.tv.display (contents & channels) and intent.navigation.section_show (sections, apps…)
BR: intent.tv.display (channels), intent.tv.app_open (apps/third party platforms) and intent.navigation.section_show (sections)

By breaking command: EXIT (“Salir”, “Abandonar”, “Quiero irme ya”, “Eu quero sair daqui”, “Terminei”)
When the user clicks on one of the contents that appears on the screen in response to their request

LLM output format

In the output format, the identified categories, their score, and the text for the user will be included without exceeding the 115-character limit. The number of tags will vary depending on different cases:

Returning 0 when it is identified that the user wants to leave the conversation (breaking command)
Returning 1 when the category mentioned by the user matches one from the available list
Returning 3 for cases such as:

No categories are identified in the user’s request
The user wants to see the full list of categories or any recommendation in general

The list of thematic categories is configured within the field tags in the preset (in args) with a format of comma-separated items enclosed in square brackets ([tag1, tag2, tag3, tag4…]).

Examples:

Input	Output
“Suficiente, quiero salir”	`{"name": "llm_conversational_search", "arguments": {"command": "EXIT"}`
“Me apetece ver algo de amor”	`{"name": "llm_conversational_search", "arguments": {"text": "Aquí tienes algo sobre la temática: Amor", "command": "", "tags":[{ "tag": "Amor", "score": 0.98 }]}}`
“Me gustaría ver contenido de aviones”	`{"name": "llm_conversational_search", "arguments": {"text": "No he encontrado nada relacionado pero puedes explorar categorías como: Épico, Bélico o Crimen fallido", "command": "", "tags":[{ "tag": "Épico", "score": 0.98 }, { "tag": "Bélico", "score": 0.92 }, { "tag": "Crimen fallido", "score": 0.89}]}}`
“¿Qué categorías tienes disponibles?”	`{"name": "llm_conversational_search", "arguments": {"text": "Estas son algunas de las categorías que puedes explorar: Amor, Crisis de pareja o Reencuentros amorosos","command": "", "tags":[{ "tag": "Amor", "score": 0.93 }, { "tag": "Crisis de pareja", "score": 0.87 }, { "tag": "Reencuentros amorosos", "score": 0.77 }]}}`

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

AI configuration by OB

The following information applies to both Spain and Brazil; there will not be separate blocks for each OB.

Introduction

LLMs UCs are configured and located within the aura-config-provision repository. See the TV LLM Search Preset (notice that in the url, the field [ob] needs to be replaced by [es] or [br]) LLM use cases have been designed to provide a fallback mechanism for user requests that currently return a ‘None’ intent or zero results, offering a second chance for classification and response generation.

Intent

intent.tv.recommend: This intent will be returned when the user seeks TV content recommendations and explicitly refers to entities such as genres, subgenres, titles, sports-related content, directors, or actors. Example: “¿Qué documentales me recomiendas?”.

Entities

Three different entities can be returned if mentioned by the user:

entities.channel: This entity will be returned whenever there is a mention of a TV channel or a third-party TV platform in the user’s request.
entities.person: This entity will be returned whenever a proper name belonging to an actor, actress, director, driver, boxer, or other public figure is mentioned in the user’s request.
entities.query: This entity will be returned when any relevant information for performing a search on the platform is mentioned in the user’s request. (e.g., TV titles, genres, subgenres)

LLM output format

Examples:

Input	Output
“Recomiéndame algún documental sobre historia”	`{"intent": "intent.tv.recommend", "entities.query": "documental\|\|historia"}`
“¿Qué películas de acción me recomiendas?”	`{"intent": "intent.tv.recommend", "entities.query": "películas\|\|acción"}`

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

AI configuration by OB

The following information applies to both Spain and Brazil; there will not be separate blocks for each OB.

Introduction

Intents

Once the triage has classified the user’s petition under the intent intent.tv.retrieve, the user’s petition can be further classified into the following intents or UCs:

intent.tv.recommend: This intent will be returned when the user seeks TV content recommendations and explicitly refers to entities such as genres, subgenres, titles, sports-related content, directors, or actors. Example: “¿Qué documentales me recomiendas?”.
intent.tv.retrieve: This intent will be returned when the user is searching for either generic or specific TV content. Example: “Muestra películas dirigidas por Quentin Tarantino”.
intent.tv.none: This intent will be returned for any request that is not covered by the two use cases outlined above. Examples: “Cambia a Telecinco”, “Avanza 5 minutos”, “Pon una alarma”.

Entities

Three different entities can be returned if mentioned by the user:

entities.channel: This entity will be returned whenever there is a mention of a TV channel or a third-party TV platform in the user’s request.
entities.person: This entity will be returned whenever a proper name belonging to an actor, actress, director, driver, boxer, or other public figure is mentioned in the user’s request.
entities.query: This entity will be returned when any relevant information for performing a search on the platform is mentioned in the user’s request. (e.g., TV titles, genres, subgenres)

LLM output format

In the output format, the only mandatory field is that of the intent. Entities are optional, as they may or may not be mentioned in the user’s request. If multiple values are present for an entity type, values must be separated by a double pipe “||”.

Examples:

Input	Output
“Pon películas de Netflix de Almodovar y Tarantino”	`{"intent": "intent.tv.retrieve", "entities.channel": "Netflix", "entities.person": "Almodovar\|\|Tarantino", "entities.query": "películas"}`
“Pon la película Titanic”	`{"intent": "intent.tv.retrieve", "entities.query": "película\|\|Titanic"}`
“Pon antena 3”	`{"intent": "intent.tv.none", "entities.channel": "antena 3"}`
“¿Qué películas de acción me recomiendas?”	`{"intent": "intent.tv.recommend", "entities.query": "películas\|\|acción"}`

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

AI configuration by OB

The following information applies to both Spain and Brazil; there will not be separate blocks for each OB.

Introduction

LLMs UCs are configured and located within the aura-config-provision repository. See the TV LLM Triage Preset (notice that in the url, the field [ob] needs to be replaced by [es] or [br]) LLM use cases have been designed to provide a fallback mechanism for user requests that currently return a ‘None’ intent or zero results, offering a second chance for classification and response generation.

Intents

This UC can classify the user’s query into three different intents:

intent.tv.retrieve: This intent will be returned when the user is searching for either generic or specific TV content or when the user seeks TV content recommendations and explicitly refers to TV content entities. Examples: “Muestra películas dirigidas por Quentin Tarantino”, “¿Qué documentales me recomiendas?”.
intent.conversational.search: This intent will be returned for user queries involving generic searches or advanced suggestions that do not explicitly reference TV content entities. Examples: “Recomendaciones avanzadas”, “Quiero ver algo en la tele”.
intent.none: This intent will be returned for any request that is not covered by the two use cases outlined above. Examples: “Cambia a Telecinco”, “Avanza 5 minutos”, “Pon una alarma”.

LLM output format

Examples:

Input	Output
“Recomendador avanzado”	`{"intent": "intent.conversational.search"}`
“Pon películas de Netflix de Almodovar y Tarantino”	`{"intent": "intent.tv.retrieve"}`
“Patatas fritas”	`{"intent": "intent.none"}`

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Aura AI use cases

Discover the AI global experiences developed by Aura Platform Team for video channels

Aura AI global cases provide different experiences to Telefónica customers for video services. Those video services could be provided by two platforms (one in Spain and GVP, currently in Brazil) and through different media, like Set-Top-Box and mobile apps, Movistar Plus.

Access the left menu and discover the available AI global experiences.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

TV Custom Recommendation UC

Global use case developed by Aura Platform Team that allows users to launch a custom recommendation based on Large Language Models (LLMs), through the use of ATRIA

Introduction

The TV Custom Recommendation use case is a global experience designed and developed by Aura Global Team that allows Telefónica customers to ask Aura for a TV recommendation based on their mood and likes, using a vocal interface.

Find additional information in the following documents:

Discover the TV Custom Recommendation specifications and the particularities of the use case development.
Discover which are the resources used by the TV Custom Recommendation use case.
Find out the NLP configuration for each OB for the TV Custom Recommendation use case.

Specifications

Kernel API

In order to resolve the user’s request, Aura uses Video Contents normalized Kernel API.

Request-response model

TV custom recommendation UC is available for use in both the deprecated request-response model v1 and the current request-response model v3.

Available channels

The TV Custom Recommendation UC is available for STB channel both in Spain and in Brazil.

Custom recommendation features

Currently, the TV custom recommendation use case includes:

A conversation flow to identify the user’s likes and mood
Once all the information is captured and treated by the LLM, a search by topic is launched to the TV APIs

Use case development

The TV custom recommendation use case development includes these components:

Understanding features

TV custom recommendation use case intent: intent.tv.custom-recommendation

In order to understand users’ requests (utterances), Aura is trained with:

NLP expression to recognize the user’s utterance and detect the user’s intention.

Use case logic

Once Aura has recognized the user’s utterance based on NLP components, the use case should be resolved based on:

Aura Bot dialog:

TV Custom Recommendation is built over the tv-custom-recommendation-v1.

Use case configuration

Check the section Configuration of the TV Custom Recommendation use case in order to know the required configuration for the TV Custom Recommendation experience for each OB.

Aura – ai

Docs:

Aura Bot generic-ai-dialog-v1

Introduction

Environment variables

How it works

Docs:

Aura Bot generic-ai-dialog-v3

Introduction

Environment variables

Adding configurations for the AI Service

How it works

Steps

Sequence diagram

Docs:

TV LLM Conversational None UC

Introduction

Specifications

Request-response model

Available channels

Use case development

Understanding features

Use case logic

Docs:

TV LLM Conversational Search UC

Introduction

Specifications

Kernel API

Request-response model

Available channels

Use case development

Understanding features

Use case logic

Docs:

TV LLM Recommend UC

Introduction

Specifications

Kernel API

Request-response model

Available channels

TV LLM Recommend features

Current limitations

Use case development

Understanding features

Use case logic

Response

Docs:

TV LLM Search UC

Introduction

Specifications

Kernel API

Request-response model

Available channels

TV LLM Search features

Current limitations

Use case development

Understanding features

Use case logic

Response

Docs:

TV LLM Triage UC

Introduction

Docs:

Introduction

Specific values

Configuration by channel

Environment variables

How it works

Steps

Sequence diagram

Docs: TV LLM Conversational None: Dialog Configuration

Configuration of the generic-ai-dialog for TV LLM Conversational None UC

Specific configuration

Docs:

Configuration of the generic-ai-dialog for TV LLM Search UC

Specific configuration

Docs:

TV Conversational Search plugin

Introduction

Validate input request