# Mnemonic seed creation in “pseudo code”

I would like to know exactly what is going on behind the process of creating the mnemonic seed from the string of bits. For example I don't understand why in the vocabulary is only around 1624 words, I would expects full 11 bit scope (I suppose that bit string is divided into 11-bit parts and each is parsed with appropriate word).. Maybe I have wrong vocabulary, in this case please point me somewhere. Thank you in advance!

I would like to know exactly what is going on behind the process of creating the mnemonic seed from the string of bits.

The whole process is explained here:

How are monero wallets generated?

For example I don't understand why in the vocabulary is only around 1624 words, I would expects full 11 bit scope (I suppose that bit string is divided into 11-bit parts and each is parsed with appropriate word)

It's 1626 words, and a mnemonic of 24 words gives 1626^24 combinations which is slightly above 2^256.

The conversion is done in triplets where each triplet is 32 bits of information:, ie `8 * 1626^3 = 8 * 2^32.0013 = 2^256.01`

It's really a conversion between a base1626 number and a base16 number (or backwards), where each word can be though of as a label for some number between 0 and 1625.

• My pleasure and welcome to StackExchange. If you're happy with the answer, please click on the "accept answer" :) – JollyMort Jul 17 '17 at 11:18