There is no audio for the Swahili lessons!
I want to learn Swahili, and I want to pronounce everything correctly, but how am I supposed to do that when it doesn't tell me how words are pronounced? Now, I did look up a key online to see how they typically pronounce each letter, but it would still be helpful if Duolingo added audio like in all the other language lessons. Is it because of its status as in beta and not stable? Swahili has been in beta for too long, it's about time it was made stable.
I've got the first 10 chapters recorded. Perhaps that will help with pronouncing some words
It was originally supposed to be released with audio but it had to put in beta without due to time constraints. Some courses don’t always have audio with the sentences; Ukrainian has some audio but it’s so quiet that there might as well not be any. The nice thing with Swahili is that it’s completely phonetic, even with double-letter sounds like “ny” (like Spanish ñ).
"The nice thing with Swahili is that it’s completely phonetic, even with double-letter sounds like “ny” (like Spanish ñ)."
Please forgive my ignorance, but what does "phonetic" mean in the sentence above? That it is easy to pronounce and listen to without having printed letters sounding different (like in English "phone" is pronounced "fon"?
"Phonetic" means that a word/the words in a language are spelled as one would expect from the sound of the word and its respective letters. So German is considered a phonetic language because every letter in every word is pronounced, and all letters have only one sound. English, by contrast, is about as non-phonetic as is humanly possible, a fact mocked by the word "ghoti" which is either silent, or pronounced in the same way as "fish". In a similar vein, the four-letter combination "ough" has nine different possible pronunciations.
To me the term is ambiguous (which I think might have been Mr.Superman's point). It's not clear if if means "written as it's pronounced"; "pronounced as it's written" (note that these aren't the same); precise one letter to one sound (er, phoneme?) bijection; or all information needed for phonetic rendering is encoded in writing by means of a one letter one sound bijection (the distinction here would be on stress in languages where it's variable). I am not entirely sure which specific criterion attains for Swahili or if there are any relevant allophones that would complicate pronunciation but that even under one-to-one letter-phoneme correspondence wouldn't have to be represented.
For instance, German meets some of these criteria, but fails one letter one sound (among others). For instance "Buch" requires four letters but only contains three phonemes.
The weird thing is that they announced months ago that the audio had actually been recorded.
However, the audio has actually disappeared for the web version of the Guaraní course, so there may be some technical issues with recorded audio (although I haven't come across such problems in the Hungarian or Ukrainian courses).
Courses are made stable when they reach threshold of average error reports per user per day (or something like that). It's not an issue of time, it's an issue of fixing bugs. Duolingo adding audio is very independent of the course's status: some launch into beta with it, others get it added during beta, and changes are made to audio even after stability.
Although the Duolingo course would be better with audio to complement the sentences, other resources should be used anyway. My Bible app has audio for Swahili, so I read, speak, listen, rewind, etc. There are probably other similar Internet resources that would have a similar function.
Yep, we might know more than the teacher does, and he does have a strong accent. (He's from Cyprus.)
"Language Transfer is the name given to the project and platform through which Mihalis Eleftheriou shares his work, and doesn't -yet- refer to any fixed team of individuals."