The short answer

Arabic's hardest sounds are made in the throat, and English has nothing to point at for any of them. You cannot learn ʿayn from a description, and a text transcript cannot tell you whether you produced one, because it records the word it thinks you meant. A tutor that receives your audio directly can hear what you actually did. We teach Modern Standard Arabic, which is a deliberate choice with real trade-offs, explained below.

Arabic has a reputation for difficulty that is mostly attached to the wrong things. The script is an alphabet and takes a couple of weeks. The grammar is systematic and beautiful, once the root system clicks.

What is genuinely hard is the sound inventory, and it is hard in a specific way: several Arabic consonants are made in parts of the vocal tract English never uses contrastively, so there is no English sound to adjust toward and no word to point at.

Why this is an audio problem

You cannot learn ʿayn from a written description. People have been trying for centuries and the descriptions are all accurate and all useless in isolation, because the instruction “constrict your pharynx” does not connect to anything you can already feel.

What works is hearing it, attempting it, and being told immediately whether that was it. Which requires something that can hear you.

Your audio goes to the model directly rather than being transcribed into text first. That is what makes the difference here, because speech recognition is built to return the word you probably meant. Say a plain a where an ʿayn belongs and the transcript will frequently hand the tutor the correct word anyway. The tutor sees a perfect sentence and says nothing.

Modern Standard Arabic, and what that means for you

We teach Fusha: the formal register of news, books and pan-Arab media, with the shared grammar and vocabulary that works across the Arab world.

This is a real trade-off and it deserves stating plainly. Nobody’s mother speaks Fusha at home. If your goal is to chat in a café in Cairo or Beirut, a dialect is what you actually need, and Fusha will make you sound like a newsreader in a bakery.

What Fusha gives you is reach and a foundation. It is understood everywhere, it unlocks written Arabic and media entirely, and every dialect is a shorter step from it than from nothing. If you want Egyptian or Levantine specifically, be aware that this is the wrong starting product for you today.

Script, and the reading underneath

The tutor writes in Arabic script, and a Latin reading of your own speech is shown underneath by default, which can be switched off in a moment.

Worth knowing that short vowels are not normally written in Arabic at all, so reading fluently depends on knowing the word already. That is a real reason to build the spoken side early rather than treating reading as the foundation.

What a session is

Unscripted conversation with something you can interrupt, backed by an authored CEFR syllabus of 251 items from A1.1 to C1, of which 105 are capabilities: concrete things you can do rather than grammar points. The pronunciation patterns above are five of twenty authored for Arabic specifically.

Travel

Travel Mode changes the app for the length of a trip: rehearsal before you go, one real-world mission a day while you are there set in the city you are in, and a debrief with your tutor afterwards. Bear the dialect point in mind here too: Fusha will be understood, and it will be noticed.

The five Arabic (MSA) tutors

Every language has five, each from a different place and each sounding like it. You can switch whenever you like, and the app remembers where you are regardless of who you talk to.

YaraCairo
Fusha spoken with the clear, broadcast-style enunciation common to Cairo-trained presenters.
KhalidRiyadh
Fusha with the steady, deliberate cadence of an educated Gulf speaker.
LaylaBeirut
Fusha carried with the bright, articulate clarity of an educated Levantine speaker.
TariqBaghdad
Fusha with the rich, resonant cadence of an educated Iraqi speaker.
AnisTunis
Fusha with the crisp, lightly French-inflected clarity of an educated Maghrebi speaker.

What English speakers usually find hard

These are drawn from the pronunciation patterns the tutor listens for in Arabic (MSA). Because it hears your voice rather than a transcript of it, they are things it can actually notice and work on with you.

ʿayn, the sound with no English equivalent whatsoever: A voiced constriction deep in the pharynx, as if gently tightening the throat while voicing a vowel. It is a full consonant, not a vowel and not a glottal stop, and it appears in the word Arabic itself. English speakers substitute a plain vowel or skip it, which changes the word.
Three different h sounds: Ordinary هـ is close to an English h. ح is a strong breathy h made deep in the throat, like fogging a mirror with the throat constricted. خ is the raspy one, as in Scottish loch. حال, هال and خال are three different words, and English hears one sound.
The emphatics, which darken everything around them: ص ض ط ظ are pronounced with the back of the tongue raised toward the throat, and they do not just sound heavy themselves, they pull neighbouring vowels darker. The a in طالب is a different a from the one in تاجر. That is why producing the consonant alone never sounds right.
Vowel length and doubled consonants are both meaningful: Long vowels are held roughly twice as long and the difference is contrastive: جَمَل is a camel and جَمال is beauty. A doubled consonant is genuinely held, and whole verb meanings ride on it: دَرَسَ is he studied, دَرَّسَ is he taught.
The l of the article disappears before half the alphabet: الـ is always written al-, but before a sun letter the l is not pronounced and the following consonant doubles instead. الشمس is written al-shams and said ash-shams. Before moon letters it stays. Nothing in the spelling changes.