What is Alexa?
Alexa is the name of the voice-based virtual assistant developed by Amazon. It’s the name of artificial intelligence technology that is used in Amazon’s Echo line, which includes several products.
Alexa was launched in 2015 in the United States and, in October 2019, Amazon started to offer the Brazilian Portuguese version of Alexa. In addition to the common functions we know available natively on Alexa (asking how the weather is, commute time, among others), it is possible to include new skills (Skills) of developers and companies with various content, customizing Alexa according to the characteristics of the user.
What are Alexa Skills?
A Skill is an application developed for the Alexa platform that makes the device increasingly intelligent. Under the hood, Amazon’s servers perform the most complicated tasks and return a run response to Alexa. The entire system is stored in the cloud to avoid problems due to lack of space.
Alexa Skills works as applications that respond to voice-driven events. New Skills can be created to customize or create new products and services.
How can Alexa be useful in the agribusiness environment?
At first, thinking about associating Alexa with agribusiness seemed to be meaningless. However, by making voice assistant connections with Machine Learning and Artificial Intelligence – the development of personal assistants itself is based on the use of AI – it is possible to assist agribusiness in a very practical way.
Some ideas for the help that the Alexa platform can generate for agribusiness:
- Alexa as a news assistant for agribusiness information:
One of the uses of a virtual assistant would be to provide agribusiness-relevant news information. Specialized agricultural media – such as newspapers, channels and online publications – can develop Skills so that their readers and subscribers have differentiated access to content produced, such as weather or even agricultural prices.
Alexa can read the most important titles of the day and search for specific information through filters. So, the countryman can get information by talking to Alexa in a simple and practical way – they can even start the day with Alexa by passing this information while having coffee, for example. This type of Skill already exists in English – Successful Farming launched its Skill for Alexa in 2017 – and it is only a matter of time before it is adopted by the agricultural media in Brazil.
- Acting as a virtual assistant for agribusiness related companies
Alexa can also help agricultural corporations through Skills that enable consumers to get information about the products offered by the organization. Data such as dosage, product validity and much more information can be made available through Product Skill to facilitate communication with farmers.
For example, a farmer who is preparing the dosage of defensives to be used, you could simply request:
Agricultor: “Alexa, what is the dose of product X in a thousand liters of water?”
Alexa can look up the dosage in the product information X and make the calculation for the dosage in a thousand liters and deliver this data very quickly and accurately. This way, the farmer does not need to search for the information and make the calculation independently, he already has the information in a fast and simple way.
- Acting via voice command to enable/disable equipment and services via the Internet of Things
Amazon announced an update from Alexa related to IoT (Internet of Things): Alexa Voice Service. This service made the voice assistant available to hardware manufacturers to facilitate IoT use on other devices through Alexa.
With this new service, it is possible to create new devices from different manufacturers, where the hardware may be able to receive voice commands, for example, to turn on / off different systems.
In agribusiness, the device that understands Alexa‘s voice commands could be used to turn on or off-field equipment. Example:
Farmer: “Alexa, turn on the greenhouse lights X.”
The device created would understand the voice command and would be able to turn off an electrical device – in this case, the greenhouse lights X. The new hardware could also include voice command and send communication to other devices with the information you wish to pass. There would be no major impact on the service itself, but it would allow you to perform tasks with a simple voice command.
- Decision-making aid for fieldwork
Imagine an Alexa Skill that, through Machine Learning and Artificial Intelligence codes, allows a farmer to get answers to complex situations. Through data processing, voice assistants like Alexa have the potential to assist the farmer in decision making by identifying or even warning about variables that he is not yet considering at the time of the task.
For example, the farmer, when thinking of spraying some area, has many other variables to consider:
- Temperature at the time of application;
- Probability of rain;
- Wind speed
- Machine availability;
- Availability of personnel;
- Proper execution of the task among many others.
A question such as “Alexa, is it going to rain in the afternoon?” May only receive a yes / no answer, while a well-developed Skill could bring other variables when answering the question:
Farmer: “Alexa, do you recommend spraying insecticide on the x plot?”
Alexa: “I do not recommend spraying now, due to the excessive wind at the moment. Air humidity is also low.”
In this case, upon receiving the question, Alexa needs to be able to understand what “gleba x” means to this producer and the system needs to identify the location of this piece of land. With this data, the system can fetch weather forecast information for the location. From this information, Alexa can send a response that helps the producer, considering the different factors that affect the issue.
Certainly, there is still a good way to go until it is possible to generate assertive information to answer such complex questions, given the enormous amounts of variables that can influence the answer. Response time and processing are certainly still bottlenecks for the development of voice assistants in agribusiness tasks, for example.
However, with a well-developed Artificial Intelligence – which considers different factors linked to agricultural production, applying historical production data and sensor information, and connects them to climate factors – it would be possible to create a Skill that can assist in agribusiness with a well-developed Artificial Intelligence – which considers different factors linked to agricultural production, applying historical production data and sensor information, and connects them to climate factors – it would be possible to create a Skill that can assist in agribusiness decision making, making -the most simple, informed and accurate.
Venturus and Alexa‘s development environment
Venturus is able to develop Skills for voice assistants like Amazon Alexa. At the launch of Alexa in Portuguese and already in the launch of skills in Portuguese, Venturus has already developed some of the Skills.
In addition, Venturus development team has participated in projects in which Alexa Skills are developed for devices and smartphones.
“Alexa, who could help us in developing Skills for Agro?”
At first, speech recognition technologies seem to have no connection to agribusiness: However, most digital systems being developed for the farmer add a lot of information relevant to agribusiness. The problem is that this information is not always structured naturally or in a way that the countryman easily extracts the data he needs.
Speech recognition technologies, such as Alexa, are changing the way we relate to companies or even equipment in many different ways. Interpreting voice messages is not simple – messages are naturally ambiguous, an intrinsic feature of human language. However, the correct interpretation of voice messages by machines has the power to improve processes and indicate different paths for technology and business.
The countryman even has some knowledge of computer systems, but generally does not fit the “nerd” stereotype — that is, he has no in-depth knowledge of systems. Often the farmer must enter a system to define the best sequence of field tasks to perform on that day. Many of the implemented systems provide this kind of help, but to get this data, it is necessary to navigate through several different screens, perform database search sequences and other information to be extracted can be quite laborious.
Voice recognition assistants can facilitate farmer interaction with obtaining the most diverse agricultural production variables. The farmer could, for example, obtain important information to make decisions about his production by talking with his assistant early in the morning, while having coffee, for example.
Given this scenario, features that ease the interaction of the rural man with the most powerful and intelligent systems are crucial for the farmer to extract better, assertive and more specialized information for the development of his main business: agriculture. A simple matter of voice to the system could save the farmer much effort and time and still provide valuable information for the farmer. In this way, voice assistants can be a very interesting solution, bringing ease to tasks that are not only important but inevitable in the farmer’s daily life.