interpretations of an array of digits

Consider a coding system for alphabets to integers where ‘a’ is represented as 1, ‘b’ as 2, .. ‘z’ as 26. Given an array of digits (1 to 9) as input, write a function that prints all valid interpretations of input array.

Input: {1, 2, 1}
Output: ("aba", "au", "la") 
[3 interpretations: aba(1,2,1), au(1,21), la(12,1)]

Input: {9, 1, 8}
Output: {"iah", "ir"} 
[2 interpretations: iah(9,1,8), ir(9,18)]

The idea here is string can take at-most two paths:
1. Proces single digit
2. Process two digit

#include<stdio.h>

char arr[]={' ','a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z'};

void change(int no[],int y, int p , int t ,char *s){

if(p==y){
*s='\0';
printf("%s \n",(s-(t)));
}

if(p<y){
    *s=arr[no[p]];
    change(no,y,p+1,t+1,s+1);
}

if(p+1<y){
     int x = 10*no[p]+no[p+1];
              if(x<=26){
              *s=arr[x];
    change(no,y,p+2,t+1,s+1);
              }
}
}

void main(){
int no[]={1,1,1};
int y = sizeof(no)/sizeof(no[0]);
char *s = malloc(sizeof(char)*(y+1)) ;
change(no,y,0,0,s);
}