As is tradition around this time, always on the eve of a new year, the consulting firm Paradigma Digital, specialized in disruptive technologies, has highlighted a handful of trends among which a new generation of virtual assistants stands out. The great skill of these prodigies lies in the ability of algorithms to interpret emotions. Roberto Torena, AI solutions architect at Paradigma Digital, assures that “AI now loses information in voice interaction because it only understands text, not emotions. From now on, we are going to hear more natural voices, with more intonation and feeling. , capable of laughing, whispering or speaking in slow tones”.
Paradigma Digital focuses its attention on the so-called Realtime API, an innovation that plans to “mark a before and after in the future of interaction between humans and computers.” According to experts from the firm owned by Indra“this technology foresees” marking a before and after in the future of interaction between humans and computers, since it is not only capable of processing voice in real time, but will make interactions much more personalized, with a conversational experience. more authentic and natural in which the virtual assistant understands what you say and interprets your mood.
There will also be room for the success of tools such as Synthesia and Descript, capable of “creating avatars that speak multiple languages or transform text into video, making it easier for small businesses to create personalized advertising campaigns at low cost and quickly.” In this short-term scenario, in 2025 we will witness the “use of voice generation models, such as VALL-E or NaturalSpeech 2, which will allow you to create hyper-realistic synthetic voices and clone voices with just a few seconds of audiofacilitating the creation of personalized virtual assistants and multimedia content in marketing, training or education, where AI will be able to create educational videos adapted to individual learning needs.”
Ana Gómez, from the Digital Strategy team at Paradigma Digital indicates that “when talking about AI in the world of work, it is always negatively associated with the elimination of jobs, but the creative will be more necessary than ever, only now they will also have to know how to transmit to machines.”
On the other hand, tools such as Agentic AI will also emerge, a technology that promises to revolutionize the role of AI with artificial intelligence systems that have a high degree of autonomy, capable of making decisions, planning actions and learning from experiences without constant human intervention. These systems can adapt to new data, interact with external tools and databases, and improve over time through continuous learning. lThe benefits include greater flexibility, precision in complex tasks, real-time interactions and an improved user experience, since they can handle multiple steps and processes autonomously. Óscar Ferrer, Goodly teampoints out that “we are going from a world based on AI to a native AI world, with autonomous artificial intelligences capable of making decisions, planning actions… with or without human supervision.”
Inclusive apps and accessible websites will also enjoy ample room for growth, in response to the fact that Almost 3 out of every 10 people in Spain have some type of limitation. “As we age, limitations cause the use of mobile phones to decline. Therefore, websites and apps face the challenge of complying with the entry into force of accessibility standards: privately owned goods and services that If they are new, they must be accessible from January 1, 2029, unless they are arranged or provided by the AA.PP, in which case the application is January 1, 2025. In this way, in “By 2025 we will see more apps conceived from their origin as accessible.”
Adrián Fernández, specialized in Mobility in Digital Paradigm, points out that “accessible applications make life easier for people who suffer from some type of irreversible disability, but also for those who, temporarily (an arm in a cast, for example), cannot use the devices with all the necessary conditions.” .
In the field of environmental respect, GreenOps raises its voice with its proposal innovative way of operating sustainably without compromising its growth. This is where GreenOps comes into play, a practice that will revolutionize the way we currently manage our cloud infrastructures. GreenOps goes beyond simply reducing your carbon footprint. It is about optimizing cloud resources intelligently, using clean and efficient technologies to optimize consumption and maximize performance, contributing to a more sustainable future. Patricia Prieto, Goodly team, points out that “GreenOps goes beyond simple cost efficiency, it is a whole new approach to minimizing environmental impact and doing things right from the beginning with sustainability always in mind.”
Industry Cloud Platforms will be talked about in 2025 with their “scalable solutions created in the cloud to meet the particular needs of a sector or industry.” These proposals “combine cloud services with applications, data and advanced technologies (AI, ML and analytics) adapted and personalized to the workflows, challenges, regulations and needs of each industry (health, manufacturing, retail, finance…) and with access to an ecosystem of specific solutions. ICPs will be key to transforming entire industries, increasing efficiency, personalizing services and improving the ability of companies to adapt to market changes,” they explain from Paradigma Digital. In that sense Juanma Fiz, Cloud Native Solutions Architect, assures that “each industry wants its cloud platform adapted to its needs and requirements and we are going to witness a complete transformation of industries with platforms from large providers or companies in a particular sector.”
Small Language Models (SLM) will emerge as a more efficient solution compared to GPT-4 or GPT-5
In his prospective work baptized techdencias 2025, Paradigma analyzes up to 24 trends that will be revealed each day of December. These include “the rise of Small Language Models (SLM) which will emerge as a more efficient solution compared to large language models (LLM) such as GPT-4 or GPT-5, which require enormous computational resources, and which are ideal for applications where resources are limited, such as in healthcare ( to analyze specific medical data, assist in the detection of rare diseases or in recommending treatments) or agriculture (to benefit from personalized analyzes without the need for expensive infrastructure).”
The same consultancy cites Gartner to point out that “andl 46% of AI projects remain in the pilot phase and are not fully implemented in companies due to the complexity of its integration into their operations. If we analyze by industry, according to Markets & Markets, the NLP market for education will grow 18.1% annually to 700 million dollars in 2028 and that of AI in agriculture will triple to more than 4.7 billion in 2028. And Gartner points out that by 2027 more than 70% of companies will use Industry Cloud Platforms to accelerate their businesses.”
#Virtual #assistants #capable #capturing #emotions #whispering #laughing #set #trend
