Juniper Research estimates that eight billion voice assistants will be in use by 2023. If this prediction is correct, it’s an impressive number. But in addition to the number of assistants, the quality of the interactions also plays a crucial role.
The Verge magazine got to the heart of it: “If you want to have an interesting conversation with someone – even a virtual someone – then it helps if they have a personality, including likes and interests.”
But we are still miles away from truly human and natural interaction. So what challenges is Conversational AI facing today and what does it mean for the future of voice assistants and co?
4 Challenges of Conversational AI
In order to understand how Conversational AI will develop in the future, it is worth looking at the status quo. The challenges that Conversational AI is currently facing provide interesting insights into future developments.
1. Ensuring safety
Ensuring security may sound trivial when you ask for the current weather report, but it is of much greater importance when you ask for other information. When users make a request to a voice assistant, the data transmitted must be securely processed and stored. This is particularly important when it comes to confidential data such as personal account data or medical information.
Voice assistants and chatbots must be taken seriously in companies and the high-security standards that companies define for these channels must be communicated to their customers in order to create the necessary basis of trust.
2. Understanding emotions
Voice assistants are becoming better at understanding users’ search queries semantically. If we look at understanding emotions, however, there is still some catching up to do. Not only WHAT is said, but also HOW it is said plays an important role in successful communication.
In order to master this challenge, the digital conversational agents must be trained with many different human voices so that they learn to identify different emotions and respond to them in an appropriate way.
3. Conversations in native languages
There are only 379 million people who speak English as their mother tongue, 753 million people speak it as a foreign language. If you count both groups together, only 20% of the world’s population speaks English.
One challenge for voice assistants will be to have conversations in a language other than English. The option of talking to a voice assistant in your mother tongue is crucial to reaching more people and building trust more easily. The dialects of different regions as well as cultural differences need to be considered.
Chatbots, voice assistants and AI,
stay informed with the free Onlim newsletter.
4. Simultaneous conversations
Living rooms are a popular place for chatbots. Adobe Analytics concludes that about 63% of smart speaker owners place them in their living rooms. In addition, most users use voice assistants on their smartphones (57%).
Both are surroundings, where several people could most likely have a conversation or give instructions. In this case, the voice assistant must be able to differentiate the command or question from the rest of the conversation. It also needs to be able to distinguish similar voices from each other and not confuse user accounts, thus revealing sensitive user data.
The Future of Conversational AI
In order to become really “conversational”, a lot of development is still necessary. There are crucial problems to be solved in the areas of security and communication behaviour. These are important to gain the trust of the broad population and to enable an interaction that feels like a real conversation.
However, when these challenges are mastered, Conversational AI and voice assistants will quickly spread in both the private and business environment and become an essential part of our everyday lives.
Read more about chatbots & voice assistants: