Vowel sounds are always voiced (the vocal folds buzz when saying these sounds). Consonant sounds can be either voiced or voiceless, and they come in pairs called cognates.  Why is it helpful to know about voiced and voiceless cognate sounds?  A frequent error in accented speech is the de-voicing of final consonants.  What this means is that the speaker does not use any or enough voicing (buzzing of the vocal folds) on the last sound of the word, so it sounds like the voiceless cognate.  This occurs for various reasons.  The speaker's native language may not have words that end with voiced sounds.  The speaker may be listening to American English speech, and not really hear or notice the voicing on the ends of words when listening to Americans.  Or the voiced sounds might be part of a group of consonants at the end of the word, and the challenge of producing this consonant cluster makes using correct voicing difficult.

Whatever the reason, attention to this error is important because you may sound like you are saying a different word than you are.  Your listener will likely understand based on the context, but having to interpret what you are saying puts an increased cognitive load on the listener, and they may struggle to follow what you are saying as a result.

The pairs of consonant sounds are as follows:
voiceless:  t    voiced:  d      Examples:  to, do
voiceless:  p    voiced:  b     Examples:  poor, boor
voiceless:  k    voiced:  g      Examples:  came, game
voiceless:  f    voiced: v        Examples:  fan, van
voiceless:  s    voiced:  z       Examples:  sip, zip
voiceless:  sh    voiced:  zh    Examples: mesh, measure
voiceless:  ch    voiced:  j       Examples:  choke, joke
voiceless:  th (IPA /θ/)  voiced th (IPA /ð/)  Examples:  thigh, thy

One helpful way to ensure you are pronouncing the final consonant with the correct voicing is to lengthen the vowel sound before the voiced consonant.  For example, when we say the word "bead," we hold the vowel ee for a little step down because the "d" is voiced.  When we say the word "beat" we don't hold the vowel ee for any additional length.  Listen to the pairs below and try making a longer vowel and voiced sound to end the word.  Then try some of the words in the list on your own.

beat   bead
bop    Bob
pick    pig
leaf     leave
bus     buzz
mesh    measure
rich    ridge
breath    breathe

Every accent has its own distinct character, but often one of the elements that can make you hard to understand has to do with the ends of words.  In American English, we usually pronounce sounds at end of a word, unless the word ends with a silent “e” (such as in the words name, goose, slice) or the word has been adopted into English from a foreign language (such as the words bourgeois, gourmet).  A common error found in many foreign accents is de-voicing final consonants (for example, saying “bet” instead of “bed”), saying the final sound of the word too softly or leaving it off entirely.  This error can be compounded by the speaker’s lack of confidence and desire not to call attention to his or her accent.

Practice reading aloud, and slow down enough to say all of the sounds you see in the written words.  You will feel your mouth moving more than you are used to.  Next, try having a short conversation using this same technique.  You may find that people can understand what you are saying better because you are putting the end sounds on your words.

Linguist Justyna Kozyra-Bober recently investigated the likelihood that the main characters in the FX hit “The Americans,” Russian-born spies living a 1980’s suburban American life, would speak with no trace of a Russian accent.  (Read her article in Language Magazine.)  While her conclusion was that the show’s realism is flawed by the fact that as adult emigrants, the Russians would not be capable of native proficiency in their pronunciation, we should not be discouraged by this information.  Instead, it is a good reminder that the goal of accent training is not to lose one’s accent entirely, but rather to improve pronunciation for clear communication.

Interestingly, the male lead, Matthew Rhys, is a Welsh actor, who was educated in Welsh-medium schools and the Royal Academy of Dramatic Art in London.  His ability to speak with a near-perfect American accent is a testament to what one would have to presume was a great dedication on his part to his American accent training. 

One of the most commonly mispronounced sounds in American English is the "th" sound. This is likely due to the fact that the “th” phonemes are rare sounds, not found in most European and Asian languages.  When we do not have a sound in our own language, the natural tendency to substitute a similar sound, such as “z,” “d” or “t.”  Unfortunately, this has a big impact on accent, because the “th” sound is found in 20% of the 25 words we say the most in English.  The most commonly used word, “the,” begins with this sound.  So when we mispronounce “th,” it is very noticeable in everyday speech.

Common mispronunciations of this sound involve improper tongue placement.  To make the “th” sound, the tip of the tongue needs to rest between the teeth.  Try making the “th” sound by itself.  First make the “quiet th,” as in the words “think” and “with.”  Then add voicing to make voiced “th,” as in “that,” “them” and “the.”