Notice the back two lines of the mouth, the /g/ has more contact points than /k/, that is really the only difference.
These are rough samples, not perfect samples.
Children who front their /k, g/ sounds often substitute /t, d/ for the back sounds. I have noticed that they often have very strong tongue tips, but somewhat uncoordinated muscles in regards to separating out tongue movement from jaw, lip or head movements. There are many strong opinions in the speech profession about whether this disorder would fall under articulation or phonological processes. I will take the stand on this situation as being a placement issues from my observations. If you want to read up on both sides of the issue go to ASHA.org website.
This brings us to HOW do therapists teach young clients to separate each mouth part yet teach the correct target sounds. Try having the client say a few words like " take, cat, mac, mickey," if all the Ks sound like /t/, most therapists cringe because they know they have their work cut out for them. Another hot debate is whether or not to use some oral motor exercises when teaching correct articulation. I'm with the mindset that you can use simple oral motor exercises briefly to help with placement and manner, but not to use them beyond their scope of utility.
I discovered a few good simple steps from close observation of clients and the assistance of the great palatometer displays. The amazing part was that I was able to see a client's progress from incorrect to correct placement in just three visits, (translated into two weeks) for isolated "K" and in word level. Most therapists without the palatomer technology work on velar sounds for months and months if not up to years.
Here are the steps for teaching correct /K, G/ placement with the palatometer.
1. Teach the client to keep their mouth open by saying "ah" (***watch out for jaw closure of any degree)
2. Now press a tongue depressor on their tongue tip as they keep their mouth open and say "kah or gah". (*** watch out that s/he doesn't push the depressor out, because they will). This simple exercise can be sent home with the depressor, so the child can practice the "kkkkah" sound with the parent as many times as they need to with and without the tongue depressor.
3. Next, have the client make a fricative /k/ like "kkkkah" sound with their mouth open, without moving their jaw closed, or moving their tongue tip up. (***Use a mirror, and s/he may have to hold their jaw still while keeping their mouth wide open)
4. Remember at first, the palatometer display showed that the client used his tongue tip to make the "kah" later as they can make the fricative "kkkah" the display will show no contact at all because they will try to make the "kkkah" very far back in the throat where the sensors do not touch. (this is all good !). When you see no sensors displays, this is great, it also means no tongue tip is being used to get the "kkkah" sound. (Cheer, this is progress)
5. Now you will start shaping that "kkkah" with k at the end of the word or middle of the word. (like (VCV in sack, or "ickey"), notice that their "k" dots will show on the display now, the "K" comes more forward producing a correct "k" sound, with no tongue tip display. This is the best because the client can now "see" they are producing the "K" correctly!
If you need to, you can go back through steps 1 to 3 again till the client can say "ickey", or "ick" or kick" or "ake, or "cake" use assimilation principle to help facilitate the two "k" sounds. later you can use font and back sounds in the word like pick, tick, pack, tack, but don't do this right away since they may revert back to fronting the /k, g/ to /t, d/. Wait on proceeding to harder words till the "K, G' is will established.
Three cheers for the Palatometer Technology!!!