How Voice Fared Last Year: 10 Things to Know

In this podcast about voice trends, Voicebot.ai’s Bret Kinsella talks with Milkana Brace, CEO of Jargon, Ravi Lal, CEO of Voxly Digital, and Eric Schwartz of Voicebot.ai. Here are some of the points made:

1. Distribution Growth – Voice devices are proliferating. Think beyond smart speakers to include mobile phones, cars, wearables and more. Bret thinks this ubiquity of voice is the story of the year. For example, Amazon alone launched 16 new products that are voice-enablers.

2. Changing Market Share for the Big Players – There was a shuffle in market share for the largest voice device providers. Baidu rose up the ranks to #2 behind Amazon in the global smart speaker market. Google fell to third – with a negative surprise in their 3rd quarter numbers.

3. Bets by the Larger Voice Companies – Like Google, Samsung hopes to leverage its advantage of having a sizable market share for non-smart speaker devices that offer voice. Google & Samsung have phones. Samsung has TVs and hotels. Samsung announced a third-party marketplace for Bixby in the middle of the year, so it has some work to do as it starts from the ground floor. Apple may have their own third-party marketplace ready by the end of 2020.

4. Contrast in Google & Amazon’s Voice App Approach – Google focuses more on the utility of its own voice apps, while Amazon has used superior marketing to be the leader. Amazon tends to focus more on its users consuming media & having fun with its speakers. In other words, Amazon is better at encouraging third-parties to build skills whereas Google tends to care more about improving its own voice functionality right now.

5. Amazon’s Skills Growth Slows – Amazon’s library has more than 100k skills now – but the growth rate in that library really tailed off after the first quarter of 2019. Perhaps because some third-parties built skills just to try it out – but haven’t gone back to build more. Skill quality perhaps is a better metric – but we don’t really know if quality is improving.

6. Automated Systems Still Require Human Help – Google Duplex is a good example of how people may be more interested in the product than the product’s ability allow (Duplex allows you to make a restaurant reservation without speaking to a human). Meaning that call centers with humans are still necessary to fix the inevitable errors that currently exist in automated systems that use voice. But the potential is there for these automated systems to be the next killer apps since they are an obvious use case for voice (the first round of killer apps include music, list-making, asking simple questions, weather). Voice with chatbots make a good combination to try to eliminate the cost of speaking to a human to fix errors.

7. Multimodal is Coming – The ability to use voice to supplement visual experiences (or vice versa) should grow , mostly on phones and TVs. Examples are Food Network’s recipes or watching Jeopardy (ie. seeing the questions on a screen). Voice could be the key to unlock the promise of augmented reality and virtual reality, which really haven’t taken off yet.

8. Who “Owns” Voice Within Bigger Brands Goes Mainstream – Ravi noted that within large companies, voice projects have moved from experimental to mainstream as the departments within these organizations that “own” the projects have moved from R&D to marketing, etc.

9. Many Tools Available to Ease Voice App Building – Eric noted how this year saw many developer offer easier ways to code voice apps.

10. First Companies Building Their Own Proprietary Voice Assistants – Milkana noted how a handful of companies have invested in building their own voice assistants, including BBC, Mercedes, Capital One, Oracle and Salesforce.