It’s Time to Add Voicebots to Your Mobile App
Want to provide more efficient and responsive service to your mobile customers, gain a competitive edge over your rivals, and improve your bottom line? It’s time to get serious about adding robust voice assistant capabilities to your mobile app.
Research has found that nearly two in three Americans aged 18 and older use voice assistants regularly, and supplemental research by Voicebot.ai found that 44 percent of consumers want their favorite mobile apps to include voice assistant functionality.
Voice assistants have come a long way in recent years, thanks to enhanced technology, and more enterprises are capitalizing on their promise and possibilities. But before committing to this upgrade, it’s important to do your homework and consider the pros and cons.
In competitive industries where companies handle large numbers of customer support queries, deploying voicebot technology within a mobile app can reap significant benefits in 2024.
First, it can enable round-the-clock customer support, addressing common issues promptly. The availability at all times eliminates lengthy waits on hold for the next available agent, and customers no longer need to wait until business hours to seek assistance.
“In this time of labor shortages and long hold times, speed of response is a key factor in customer loyalty,” says Amanda Robinson, director of conversational AI strategy and design at TTEC Digital. “It also offers a more seamless customer experience, allowing your mobile users—often your most engaged customer segment—to reach out for contextual support right where they need it.”
Brian Prince, founder and CEO of Top AI Tools, agrees.
“Adding voice assistants to mobile apps can significantly improve user engagement and experience. Research by PwC shows that 71 percent of people would rather use voice search than type a search query, so it stands to reason they would like the same capability in a mobile app. Voice assistants offer a more intuitive, conversational way for users to interact with apps, reducing the effort to navigate and perform tasks. This makes apps more accessible to a broader range of consumers, while also helping in multitasking,” Prince says.
Studies demonstrate that rapid, comprehensive support ranks as the highest service priority for consumers. Voice assistants can promptly address repetitive inquiries, allowing live agents to concentrate on more complex questions that necessitate a compassionate human approach.
“This tiered, human-in-the-loop approach maximizes issue resolution, and with voice assistants handling high-volume routine requests, companies can slash staffing costs. Plus, lower call volumes also improve job satisfaction for human agents, boosting retention,” Nikola Mrkšic, CEO and cofounder of PolyAI, explains. “The internal and external benefits of [voice assistants] make them valuable technology investments for companies that want to take their mobile app—and their reputation—to the next level. And who doesn’t want a reputation as the company with memorably outstanding customer service?”
Diane Johnson, head of customer experience at Sendbird, notes that, with recent advances, the now near-natural tone of voice assistants can establish positive connections.
“Customers also benefit from accessibility and the convenience of hands-free interactions when working with voice assistants, which can help the visually impaired as well as non-tech-savvy demographics. This widens an application’s reach, which should be a goal of companies,” she says.
Furthermore, voice interactions provide valuable data on user preferences and behavior, which can then be analyzed to personalize services and improve product offerings.
“This can also differentiate a company from their competitors as more tech-savvy customers may prefer voice interaction, increasing your market,” says Damian Edwards, commercial manager for Omnie AI. “As voice assistants can support multiple languages, this helps companies expand their services globally, too, and therefore reach a wider audience.”
Justin Lundy, founder and CEO of Lundy.io, is convinced that businesses that preemptively adopt voice can set a new standard in customer service.
“But those slow to adapt risk becoming obsolete in a rapidly evolving market where convenience and innovation lead,” he cautions.
SUCCESS STORIES
For proof of how impactful voice assistants can be in mobile settings, Walmart has integrated voice shopping features into its mobile app via Google Assistant and Siri, enabling customers to add items to their carts simply by speaking to the assistant. Similarly, in the travel industry, Expedia offers a voice-activated hotel booking feature within its mobile app, streamlining the booking process and elevating the user experience by facilitating a more natural, conversational interaction.
“There are many impressive real-world examples today of mobile apps with enhanced customer support services where voicebots can handle routine inquiries, bookings, and troubleshooting, allowing human agents to focus on more complex issues,” Prince says. “Banks with mobile apps can use a voice assistant to help users check their balances, transfer money, or find the nearest ATM by simply speaking to the app. Another example is in the hospitality industry, where hotels can employ voice assistants to streamline check-ins and room service orders.”
Lundy’s integration of voice assistants into its Finding Homes app enables users to conduct property searches using simple voice commands.
“A visually impaired user could request, ‘Show me homes in Boston under $500,000,’ and receive instant, tailored results,” Lundy says. “This approach mirrors the convenience seen in the Domino’s pizza voice ordering system, called Dom, but applies it to real estate, showcasing how voice technology can transform accessibility and efficiency in customer interactions.”
And in healthcare, “patients could use the voice assistant to schedule appointments, request prescription refills, or ask for information about services,” Edwards says. “A patient could say, ‘I need an appointment with Dr. Smith next week,’ and the voice assistant could respond with available times, book the appointment, and even provide pre-visit instructions.”
IMPROVED TECHNOLOGY
How did voice assistants reach this level of effectiveness within mobile apps? Robinson points to four recent innovations in particular:
- Improvements in automatic speech recognition (ASR), attaining human-level transcription levels in ideal conditions.
- Natural language understanding (NLU) advances that allow users to speak naturally rather than having to speak in cryptic commands.
- Innovations in synthesized speech or text-to-speech (TTS), including neural voices that sound incredibly natural, and voice-modeling technology that can create custom voices using only a few samples of a person’s voice.
- Embeddable voice engines that can run locally on mobile devices.
Lundy credits NLU progress as perhaps the most important factor here.
“Voice assistant technologies have advanced to understand natural language more accurately, recognize user intent, and offer personalized interactions. These improvements make them ideal for mobile apps to enhance customer service by facilitating more intuitive and efficient user experiences,” he says.
In addition, robotic-sounding interactive voice response (IVR) systems have been banished to the past thanks to much-improved conversational AI capabilities.
“These newer, intelligent voice assistants can understand speech much better than traditional frustrating IVRs that are reliant on specific keywords to move the conversation forward,” Mrkšic notes. “Additionally, many people are inclined to trust effective voice assistants more than IVRs since IVRs often fail to resolve customer needs because of their technical limitations.”
Previously, voice assistants had to follow a linear decision-making structure, assuming that the conversation would go in one particular direction.
“But now, with large language models, the assistant can context-switch, which adds to the naturalness of the conversation,” says Zohaib Ahmed, CEO and cofounder of Resemble AI.
Ian Cahill, director of innovation and data at VML, also points to major advances in AI.
“With the advent of technologies like ChatGPT, the pairing of voice and AI makes it possible to create convincing digital humans that can interact with users in a more human and meaningful way to address customer issues,” he says. “It is now possible to feel like you are being assisted by something or someone who really cares about your needs and can identify in a real way the exact problem or issue you are facing within an app.”
Voice assistants nowadays can also seamlessly connect with diverse services through APIs, enabling them to streamline tasks such as flight booking, hotel reservations, accessing data/news, and conducting payments, enhancing both efficiency and customer satisfaction.
“And increases in processing power on mobile devices means there are now toolkits that support robust speech recognition and synthesized speech without needing a network connection at all,” Robinson adds.
CHALLENGES TO OVERCOME
Still, impediments remain that threaten to block the path to swifter progress and adoption of voice assistants within mobile apps, experts point out.
“Challenges do persist, including occasional inaccuracies in understanding complex questions and interoperability issues, which require ongoing advancements in natural language processing and integration capabilities,” Johnson says.
Another major hurdle? “Humans haven’t fully learned how to be comfortable speaking with these virtual beings,” Dave Hoekstra, product evangelist at Calabrio, says. “The primary limitation that VAs are facing is that people still don’t trust them, so how can we increase the ability for these assistants to help people with few to no barriers?”
There’s also the problem of noisy environments and the fact that current NLU technology has difficulty maintaining conversational context the way a human conversational partner would.
“This means a voicebot may not remember something mentioned a few minutes ago, may not be able to retrace its steps back if you change the subject, and so forth,” Robinson says.
And then, there are issues with understanding diverse accents, dialects, and the nuances of human speech, which can sometimes lead to incorrect responses. Privacy concerns remain, too, as users might be wary of their conversations being recorded or analyzed.
“Companies need to think very carefully about data privacy and security measures. They must be clear and intentional in their approach to safeguard sensitive customer information,” Johnson advises.
IMPLEMENTATION IDEAS
Eager to make your mobile application voice-enabled? Lundy recommends these seven best practices:
- Select an appropriate voice technology platform that aligns with your application’s specific requirements, including language support and customization options. Google’s Dialogflow, Amazon Alexa Skills Kit, and Apple’s Sirikit offer comprehensive tools for developing sophisticated voice functionalities.
- Design for a conversational user experience. “The goal is to create a user experience that mimics a natural, intuitive conversation. This involves crafting voice interactions that are clear and responsive, guiding users efficiently through their tasks,” Lundy says.
- Ensure user privacy and security. Emphasize the importance of safeguarding user data by establishing strong privacy protocols. It’s crucial to openly communicate how voice data will be used and stored; obtain consent; and encrypt the data.
- Focus on accessibility. “Make the voice assistant accessible to a diverse user base, including those with visual impairments or other disabilities,” Lundy urges. “This means ensuring all app functions are voice-command-compatible and integrating with accessibility tools like screen readers.”
- Iterate based on testing and feedback. Perform comprehensive testing across diverse user demographics to detect issues related to voice recognition, especially concerning accents and languages. Use feedback to enhance and refine the voice interaction experience.
- Educate users on voice features by providing clear, accessible instructions on how to use the voice commands within your app. A concise tutorial or assistance section can greatly improve user engagement and familiarity with these functionalities.
- Monitor and adapt to user feedback by continuously assessing how the voicebot is performing and staying responsive to user feedback. Continuous analysis will guide essential adaptations and advancements to ensure the service remains pertinent and user-friendly.
Whether selecting packaged solutions from providers like Amazon Web Services or developing a custom stack using private or open-source APIs, your business “must prioritize scalability, reliability, and security throughout the integration process,” Johnson insists. “By choosing the right provider, companies can be sure that they have access to the necessary tools to easily add voice assistants to their platforms in ways that truly enrich customer experiences.”
If your company has transitioned its contact center to the cloud and has already invested in voice automation, many platforms let users integrate voicebot functionality into their mobile applications.
“Note that, in this model, the ears and mouth of the voicebot will reside in the cloud, and the app will send speech data back and forth for every turn of the conversation,” Robinson cautions. “If cloud support is not attractive or raises concerns, such as data privacy, response speed, or network availability, some options allow embedding the speech recognition and synthesis so that everything stays right on the local device.”
Prior to committing to voicebot integration, set realistic expectations and avoid common mistakes that some enterprises make.
“You can likely anticipate voice assistants to significantly enhance user engagement and satisfaction by offering an intuitive and efficient way to interact with your app. And you can look forward to improved customer service capabilities, with faster response times and the ability to handle a high volume of routine inquiries automatically,” Prince says. “However, be prepared for the challenges of integrating this sophisticated tech, including ongoing training and refining the voice assistant to better respond to customers. Expect gradual implementation, continuous improvement based on user feedback, and balancing automation with human oversight.”
Robinson warns against adopting an ‘If you build it, they will come’ mindset. “Users will always choose the modality that best meets their needs. If you add voice support without talking to actual users and aligning the feature with their needs and context, there’s a risk that they will not engage with it as much as you expected.”
Just as you wouldn’t expect a human customer service agent to excel at 100 percent capacity on day one, voicebots also need onboarding, training, and continual monitoring and adjustments to enhance their performance gradually.
“Plan to monitor and make adjustments for at least six months following the launch, ideally longer,” Robinson advises.
Above all, recognize that these assistants will never completely replace human customer service representatives. While intelligent voice assistants can proficiently manage many routine inquiries and notably reduce call volume, specific issues require a nuanced understanding and a human touch that voice assistants still cannot offer.
“No matter how sophisticated voice AI becomes, companies and apps should still be transparent with users,” Cahill says. “Voice assistants and AI will not replace humans but can be tasked to handle more common use cases and baseline needs. As users become more comfortable with these assistants and frankly prefer them based on the quality of the responses, they will become a mainstay in all customer support operations. But to get there, we must be aware of the types of queries that are still best suited for the softness of a human response.”
Erik J. Martin is a Chicago area-based freelance writer and public relations expert whose articles have been featured in AARP The Magazine, Reader’s Digest, The Costco Connection, and other publications. He often writes on topics related to real estate, business, technology, healthcare, insurance, and entertainment. He also publishes several blogs, including martinspiration.com and cineversegroup.com.