In my personal opinion, Apple appears to take a global approach to building its language footprint, rather than simply prioritising languages according to the number of speakers in the world.
Instead of concentrating heavily on one country because it has a very large population and multiple major languages, Apple seems to be trying to establish Siri across as many countries, regions and language communities as reasonably possible. The objective therefore appears to be broader geographical and linguistic coverage—not necessarily completing every major language in one country before moving on to another part of the world.
This helps explain why some languages with relatively smaller speaker populations have full Siri support, while some languages with hundreds of millions of speakers do not yet have a complete Siri experience.
Vietnamese is a good example. Bengali has a larger worldwide speaker population, yet Vietnamese is included among the languages supported by Apple's newer Siri experience. This illustrates that speaker population alone cannot explain Apple's language priorities.
India provides an even more interesting example. It has enormous linguistic diversity, with languages such as Hindi, Bengali, Marathi, Tamil, Telugu, Gujarati, Kannada, Malayalam and Punjabi having substantial speaker populations. Rather than providing a complete standalone Siri experience in every one of these languages, Apple has developed multilingual speech recognition for India, allowing users to mix English with several Indian languages.
Therefore, Apple's language strategy can reasonably be viewed as an effort to build a global language footprint while balancing geographical reach, regional demand, technical feasibility, language complexity, engineering resources and the ability to deliver a consistent Siri experience.
So, the fact that Bengali is one of the world's most widely spoken languages does not necessarily mean it would automatically receive priority over Vietnamese, Danish, Norwegian or Swedish. Apple appears to be looking at the worldwide Siri language footprint as a whole, rather than simply selecting languages according to population rankings within individual countries.
In that sense, Apple's approach seems less like “Which country has the most speakers?” and more like “How broadly can we extend Siri's language footprint across the world while maintaining a reliable experience?”
