Speech and voice recognition technologies are major market trends, but what about their actual use by brands and consumers? Speech and voice technologies have a hard time becoming a part of people’s habits since they do not consider the human factor. Word error, nuanced communications, understanding problems… these are some of the reasons why assistive technology such as speech recognition or vocal synthesis are not fully adopted by end-customers and so, are still a must-have for call centers. However, these technologies are growing in use and therefore have not said their last word. Let’s understand why.
When customers contact brands today, they want a quality experience that also saves them time.
Many current technologies assist customers in this manner, from IVR menus that direct customers to the most qualified agents as quickly as possible, to AI-powered chatbots on instant messaging apps that help customers find quick answers through self-service. While customers often use written text or touch tone technology to communicate through these communication methods, other call center technologies focus on voice recognition technology.
While automated speech recognition and speech synthesis technologies offer numerous benefits for both companies and customers—from reduced costs for companies to saved time and greater self-service for customers—these technologies have not been as widely adopted as once expected.
Despite its potential benefits, the adoption of speech recognition technology has been much slower than anticipated. In 2000, Gartner predicted that 30% of customer service departments would be using this technology by 2003. However, global adoption ended up being much slower than projected, despite the fact that speech recognition technology had already been around for years. It is possible that companies found the technology to be too lacking in advanced features—making it unreliable in a customer service context. For example, speech recognition technology does not fully take into account the needs of all customer profiles, such as deaf-mute customers or customers with accents or languages that the technology cannot identify. It does not take into consideration disabilities and how people speak. In addition, factors such as background noise (for example, street sounds when a customer is out) can make it difficult or even impossible to complete a successful call. The technology would likely ask the customer to repeat information, leading to frustration and customer dissatisfaction. For companies, it is also difficult to configure speech recognition technology. Identifying numerous phonemes, or the sounds that make up language, to program such technology is both long and difficult.
Still, the adoption of this technology is picking up again today. Increased smartphone adoption, advancements in machine learning and speech recognition technology, and the promise of lower costs for companies—as it is cheaper to use this technology than assign human agents to every role—is fueling the market’s growth in the customer service, healthcare, and financial sectors.
Many companies find that text-to-speech technology is more relevant for their call centers. This speech synthesis technology converts written text to spoken language, once again using a synthetic yet realistic voice. The technology is easily customizable, as it can feature a male or female voice as well as a variety of languages and accents. This capability therefore allows companies to match certain voices to specific customer profiles or a specific brand image. It can help personalize the customer experience without a need for human agents. While it is often used to assist customers with inbound calls, it may also be used for outbound calls when delivering important messages to customers (such as a scheduled power outage or a reminder that a bill is available for viewing).
More importantly, TTS helps drive self-service. It can search for specific words in a customer’s profile contained in a CRM database, and then provide a response to the customer’s request using the data it has obtained. For example, TTS can search a CRM database to give a customer a bank account balance or scheduled delivery date. Self-service is essential to great customer experience, as it allows customers to find quick support 24/7 without the need for human assistance. Ultimately, self-service saves customers time and increases their satisfaction. This also means that human agents have more time to manage tasks that do need their intervention, such as communicating with customers who have complex cases. The more questions that are answered through self-service, the less customers will need to contact your call center on channels such as voice that cost more to operate. However, it is important to note that TTS works best for simple requests—the technology cannot work well with more complex tasks that go beyond verifying concrete data such as dates, balances, etc.
While speech recognition and speech synthesis technologies are slowly gaining popularity, they are not the norm just yet. As these technologies become more advanced and therefore more reliable and easier to use, companies and customers alike are more likely to adopt them.
