This episode currently has no reviews.
Submit ReviewMachine language processing is a key part in the development of voice technology, smart speakers, and voice marketing around the world. Let's start with the basics: how complex is it to process different languages?
First, English is the most information dense language per sound at 1.08 followed by French at 0.99. This means we can convey more information in fewer spoken syllables.
Source: https://www.realclearscience.com/blog/2015/06/whats_the_most_efficient_language.html
How to find the information rate: come up with a value for how much meaning is packed into each syllable, which will give you an average information density per syllable. use those values to derive an information rate.
Languages which have a lower average density of information per syllable (because they have on average longer words or more lengthy grammatical structures) compensate in effect by being spoken faster: this is why Spanish is spoken faster than English.
That said, Chinese is most information dense per written character, and per faded ankle tattoo.
Interestingly, the languages that conveyed the least amount of information per syllable, like Spanish, Japanese, and French, tended to be spoken at a faster rate. This allowed these languages (apart from Japanese) to deliver a similar amount of information compared to more meaning-dense languages.
However, it does not make English superior.
Want to learn more about speech recognition and machine language processing? Check out Carl Robinson's Voice Tech Podcast.
Further reading:
http://content.time.com/time/health/article/0,8599,2091477,00.html
https://www.quora.com/What-spoken-language-carries-the-most-information-per-sound-or-time-of-speech
Hosted on Acast. See acast.com/privacy for more information.
This episode currently has no reviews.
Submit ReviewThis episode could use a review! Have anything to say about it? Share your thoughts using the button below.
Submit Review