A PhD's Take on the Intricacies of Language & Connection
May 1, 2025
Can you share where your love for language and linguistics first began?
My love for language began at a very young age, driven by my desire to travel the world, meet people from different countries and cultures, and communicate with them. This led me to pursue an undergraduate degree in Linguistics, where I quickly became captivated by the intricacies of how languages work, how they encode meaning, and the cognitive processes behind communication. This curiosity led me to pursue a PhD in Linguistics, where I deepened my interest in Construction Grammar and Frame Semantics, particularly through my research at the Federal University of Juiz de Fora and the University of Gothenburg. This journey allowed me to explore how different languages structure meaning and convey complex ideas through constructions.
How has language shaped your personal and professional life?
Language has profoundly shaped both my personal and professional life. Personally, it has enabled me to engage with different cultures, both through learning languages and living incredible experiences, connecting with people. Professionally, language is the foundation of my research. My work has focused on understanding how constructions—patterns of form and meaning—work across languages like Brazilian Portuguese, Swedish, and English. These insights have not only contributed to theoretical linguistic research but also have practical applications in natural language processing and machine translation. Finally, it has given me the tools to understand and teach language at a deeper level, particularly through my work in linguistics and language teaching.
Often when talking about translation, the core concept of linguistics and language patterns is often overlooked – as an expert in this field, why do you feel it is essential to incorporate deep linguistic learnings while developing multilingual technology?
Translation is far more than a simple word-for-word conversion. It involves understanding and preserving the intricate relationships between form, meaning, and context in different languages. My research in contrastive linguistics and construction grammar has shown that each language has its own set of constructions, which encode distinct ways of conceptualizing the world. If multilingual technology, like machine translation, does not account for these underlying linguistic patterns, it risks oversimplifying or misrepresenting meaning. Deep linguistic insights are essential for ensuring accurate translations that reflect the cultural and functional nuances of each language.
Have you used other AI translation tools before? If so, what were the biggest frustrations or issues you experienced?
Yes, I’ve used several machine translation tools in my research, such as Google Translate and other neural machine translation systems. One of the main frustrations is the lack of sensitivity to syntactic and semantic nuances. For example, machine translation often struggles with idiomatic expressions, which cannot be understood purely through the direct translation of words. Furthermore, when dealing with constructions that are highly language-specific, the output can be awkward or even unintelligible. This highlights the need for translations that go beyond surface-level word mapping and account for deeper linguistic structures, which is something I’ve worked on extensively in my research.
Did you ever find translations from other tools inaccurate or unnatural? Can you share an example where it led to confusion or miscommunication?
Machine translation tools often struggle with regional dialects and conversational tones because these aspects of language involve a rich context that goes beyond vocabulary and grammar. For instance, in Brazilian Portuguese, the use of informal speech in different contexts can drastically change the meaning of a sentence. A translation tool that doesn’t account for these variations might generate a translation that sounds formal or completely misses the tone. For example, the expression "Pega leve" can carry subtle, informal, and context-specific tones, and this nuance might not always be captured. In one context, it might be used to encourage someone to relax, while in another, it could be a playful way to tell someone to be less intense. This is where understanding constructions at a deep linguistic level is crucial.
Cultural understanding is extremely important when it comes to language – in your culture and specific dialect, what are some of the biggest miscommunications that may arise when it comes to using machine translation for communication?
In Brazilian Portuguese, there are many expressions that are deeply rooted in cultural context. For instance, the expression “dar uma mãozinha” (literally “to give a little hand”) is a colloquial way of saying “to help someone,” but this would be confusing if translated literally into English. Machine translation systems often overlook these cultural meanings, leading to a loss of the intended message.
The first time you used Aavaaz, what was your immediate reaction? Did anything surprise you?
Using Aavaaz for the first time was refreshing because it showed a clear understanding of regional dialects and conversational nuances, something that other machine translation tools often miss. I was particularly impressed by how Aavaaz managed to capture the subtleties in Brazilian Portuguese expressions, which are often challenging for other tools to translate effectively. I tried to be very informal and colloquial when using Aavaaz and it captured the contextual meaning. That really impressed me.
As a PhD and expert in the field of linguistics, how do you feel Aavaaz handles linguistic ambiguity that comes with translation?
Aavaaz seems to handle linguistic ambiguity in a sophisticated way by considering both the syntactic and semantic context of words and phrases. In my own research, I’ve shown how constructions can have different meanings depending on the context, and Aavaaz appears to account for these variations effectively. This is important because many ambiguities in language arise not from individual words but from how they interact within larger constructions. Aavaaz’s ability to address this makes it stand out compared to more conventional machine translation tools. For example, I tested the sentence “Eu tô com tanto trabalho que tô quase chutando o balde!”. The Brazilian Portuguese construction “chutar o balde” would be literally translated as “kick the bucket.” However, “kick the bucket” and “chutar o balde” have different contextual meanings. In English, “kick the bucket” means “to die,” whereas in Brazilian Portuguese, “chutar o balde” means “to give up” or “throw in the towel”—but not necessarily to give up on life. Most machine translation tools provide a literal translation. However, Aavaaz gave a very accurate translation, which was quite impressive.
How would you describe the difference between Aavaaz’s translations and the translations from other machine tools like Google or GPT?
The key difference between Aavaaz and other machine translation tools like Google Translate or GPT is that Aavaaz seems to have a deeper understanding of the linguistic constructions behind the words. Other tools often rely on direct translations of words or phrases, which can lead to awkward or incorrect renderings, especially with idiomatic expressions. Aavaaz, on the other hand, shows a more sophisticated grasp of how meaning is constructed in context. For example, my own research in contrastive linguistics demonstrates that language is much more than just a set of words—it’s about how these words fit together in meaningful ways across different languages, which Aavaaz seems to understand better than many other systems.
Aavaaz specializes in translating colloquial and regional dialects. What you noticed in terms of how it specifically understands your language and cultural nuances?
Aavaaz’s ability to handle colloquial and regional dialects is impressive. As someone who has worked on contrastive analysis between Brazilian Portuguese, Swedish, and English, I’ve seen firsthand how challenging it can be to capture the regional and cultural nuances in language. Aavaaz did an excellent job of understanding these subtleties. For example, in Brazilian Portuguese, certain expressions or idioms can only be understood within specific cultural contexts. Aavaaz managed to correctly interpret these constructions without losing their regional flavor. This is a major advantage over many other translation tools that often struggle with such nuances.
On a percent score out of 100%, how would you rate your trust in Aavaaz being used for communication with your loved one in a critical situation?
Given its ability to accurately capture linguistic nuance and context, I would rate my trust in Aavaaz at 95%. As a linguist, I understand the complexity of human communication, especially in critical situations where clarity is essential. Aavaaz seems to offer a high level of reliability when it comes to translating not just words, but their underlying meaning, which is vital for clear communication. While no tool is perfect, Aavaaz demonstrates a strong potential for use in important conversations, even in high-stakes situations.
Now, what about the same for if Google Translate is used?
If Google Translate were used, I would rate my trust at around 60%. While it is useful for basic translations, it often misses the cultural and contextual subtleties that are crucial in a critical conversation. In my research, I’ve demonstrated how construction-based approaches allow for a deeper understanding of meaning, which Google Translate does not always capture. I have encountered many bad translations given by that tool. Since my language students rely too much in Google Translate, they constantly say things that doesn’t actually make sense in real contexts of use.
As a PhD and language teacher, your students look towards you with a lot of credibility and trust. Similarly, in working with Aavaaz, level of trust do you have in Aavaaz supporting your students as a reflection of you and your business?
As a language teacher and researcher, I highly value accuracy and clarity in communication. I would trust Aavaaz to support my students, as it aligns well with the linguistic principles I teach them—namely, understanding and interpreting the meaning of constructions in various contexts. Since Aavaaz takes into account the broader linguistic and cultural context, I would feel confident recommending it to my students. The tool’s ability to handle the complexities of language construction gives me confidence that it can serve as a valuable resource in their learning journey.
What real-world impact do you feel Aavaaz can create to help millions around the world feel more included and accepted through language?
Aavaaz has the potential to bridge communication gaps across languages and cultures, making the world more inclusive and connected. In my own research, I’ve seen how language can be a powerful tool for creating understanding, but also for dividing people due to misunderstandings or mistranslations. Aavaaz’s approach to translating not just the words but the underlying meaning of constructions can foster a deeper level of cross-cultural understanding. This could help people from different linguistic backgrounds communicate more effectively, reducing the barriers that often exist due to miscommunication and fostering inclusivity in global conversations.
How has Aavaaz changed the way you perceive the power of technology to advance global multilingual connection?
Aavaaz has significantly shifted my perspective on the role of technology in global multilingual communication. As a linguist, I’ve always seen language as a complex, context-dependent system, and for years, I’ve observed how traditional translation tools often fall short in capturing the nuances of meaning and context. Aavaaz, however, demonstrates that advanced technology can understand and respect these subtleties. By focusing on linguistic constructions, rather than just translating individual words, Aavaaz paves the way for more natural, context-sensitive communication. This approach holds great potential for breaking down language barriers globally, allowing people from different linguistic and cultural backgrounds to connect more authentically and effectively. It shows that technology can be a powerful force in promoting global multilingual connection, not just by translating words, but by facilitating meaningful, culturally aware exchanges.
After using Aavaaz, what would you say to someone who is skeptical about trusting AI translation, especially in sensitive situations and conversations?
I would reassure them that Aavaaz represents a significant step forward in AI-driven translation tools. As someone with a deep background in computational linguistics and the study of constructions, I can confidently say that Aavaaz handles linguistic nuance with a sophistication that most other tools don’t. In sensitive situations, it’s essential that translations not only be accurate in terms of vocabulary but also in terms of context, tone, and cultural nuances. Aavaaz’s approach to understanding these layers of meaning is much closer to how a human translator would process information, ensuring that sensitive conversations are handled with care. I would encourage them to test Aavaaz in a real-world scenario, as I’ve been impressed by how well it can maintain the integrity of meaning, especially when dealing with complex or culturally specific expressions.
If you could describe Aavaaz in just one word, what would it be and why?
“Context-aware.” The reason I would choose this word is that Aavaaz goes beyond simply translating words and instead understands and respects the broader context of language use. This context-awareness allows Aavaaz to produce translations that are not only linguistically accurate but also culturally and situationally appropriate. This is essential in ensuring that the translation captures the true meaning of what is being communicated, rather than just a surface-level conversion of words.
What are you most looking forward to seeing Aavaaz build as we grow?
I’m really excited to see how Aavaaz continues to grow, especially in fields like education and even travel – the fields I have more interest because that aligns with my students interests too. In healthcare, where clear communication is crucial, Aavaaz could revolutionize how patients and doctors from different linguistic backgrounds interact. But beyond that, I think Aavaaz could do wonders in education by helping teachers and students communicate better across languages, especially in multicultural classrooms. In the travel and tourism industry, it could help travelers break down barriers and connect with locals in ways that haven’t been possible before. I'm really looking forward to seeing Aavaaz expand into these areas and make a real difference in even more people's lives!
Dr.Laviola’s expertise and trust in Aavaaz continue to be a fueling force in the drive to build equitable healthcare experiences globally – where human conversation serves as an opportunity to connect, not a place to divide. As a PhD, teacher and business owner we are proud to have Dr.Laviola sit among our panel of worldwide language experts – each graciously supporting the relentless efforts to bring the global AAVAAZ (voice) to life.
