Learn English – American English without an accent

“Standard” American English (SAE), when used to describe accents, is identical to the term “General American English”, and means American English spoken without any clear regional dialect markers. Sociolinguists tell us that this means essentially English spoken in the north midlands region, like Iowa, but in reality it is spoken all over the country, even in places where many people speak with what others would describe as “heavy” or “thick” accents. SAE is the English used by most television and radio broadcasters—that is, those who don’t speak with an identifiable regional accent.

In truth, a “standard” American accent is defined more by what it isn’t than what it is. It does not have the Northern Cities Vowel Shift, New England non-rhoticity, features of New York English, or any of the phonological features of Southern American English.

There is some debate as to whether speakers with or without the cot-caught merger (that is, whether or not the words cot and caught are pronounced differently) are speaking SAE, but most would consider that not a defining characteristic of SAE.

You might be interested in The TELSUR Project, which is a high-granularity phonological survey of North American dialects. Note that the information presented on that site is very detailed and highly technical.

In particular, there is an overview of major dialect regions, of which they distinguish five major groups:

1. West (basically everything west of the Great Plains)
2. North Central (MN, WI, and the Dakotas)
3. Midland (most of the Great Plains and the Midwest)
4. Inland North (MI and inland NY)
5. South (um, the South)

The Northeast presents special difficulties, as it has the most diversity of any region, and isn't represented in the regions above. Here is a detail map of Northeast dialect regions.

However, there are a multitude of subtle gradations in these divisions, which can be explored by reading through the materials presented on those sites.

You might also like to look through the Harvard dialect survey results, which plot an enormous number of phonological and lexical differences across the country.