#pragma GCC optimize("Ofast")
#pragma GCC target("avx,avx2,fma")
//#pragma GCC target("sse,sse2,sse3,ssse3,sse4,popcnt,abm,mmx,tune=native")
#include <bits/stdc++.h>
using namespace std;
#define rep(i,n) for(int i=1;i<=n;++i)
#define per(i,n) for(int i=n;i>=1;--i)
#define pb push_back
#define mp make_pair
#define inf 0x3f3f3f3f
#define N 600
#define M 510000
struct NetWorkFlow {
    struct EDGE {
        int adj, w, cost, next;
    } edge[M*2];
    int gh[N], q[N], dist[N], v[N], pre[N], prev[N], top;
    int S, T;
    void addedge(int x, int y, int w, int cost) {
        edge[++top].adj = y;
        edge[top].w = w;
        edge[top].cost = cost;
        edge[top].next = gh[x];
        gh[x] = top;
        edge[++top].adj = x;
        edge[top].w = 0;
        edge[top].cost = -cost;
        edge[top].next = gh[y];
        gh[y] = top;
    }
    void clear() {
        top = 1;
        memset(gh, 0, sizeof(gh));
    }
    int spfa() {
        memset(dist, 63, sizeof(dist));
        memset(v, 0, sizeof(v));
        int head = 0, tail = 1;
        q[1] = S; v[S] = 1; dist[S] = 0;
        while (head != tail) {
            (head += 1) %= N;
            int x = q[head];
            v[x] = 0;
            for (int p=gh[x]; p; p=edge[p].next)
                if (edge[p].w && dist[x] + edge[p].cost < dist[edge[p].adj]) {
                    dist[edge[p].adj] = dist[x] + edge[p].cost;
                    pre[edge[p].adj] = x;
                    prev[edge[p].adj] = p;
                    if (!v[edge[p].adj]) {
                        v[edge[p].adj] = 1;
                        (tail += 1) %= N;
                        q[tail] = edge[p].adj;
                    }
                }
        }
        return dist[T] != inf;
    }
    int work() {
        int ans = 0,flow=0;
        while (spfa()) {
            int mx = inf;
            for (int x=T;x!=S;x=pre[x])
                mx = min(edge[prev[x]].w, mx);
            ans += dist[T] * mx; 
			flow+=mx;
            for (int x=T;x!=S;x=pre[x]) {
                edge[prev[x]].w -= mx;
                edge[prev[x]^1].w += mx;
            }
        }
        return flow;
    }
} nwf;
char ans[510];
int l[30],r[30];
int can[510][30],id[510][30];
int g,len;
char s[510],c[510];
int main()
{
	cin>>g>>len;
	for(int i=0;i<26;i++)l[i]=0,r[i]=len;
	
	for(int i=1;i<=len;i++)
	for(int j=0;j<26;j++)
		can[i][j]=1;

	rep(ii,g-1)
	{
		scanf("%s%s",s+1,c+1);
		static int cur[30];
		for(int i=0;i<26;i++)cur[i]=0;
		rep(i,len)
		if (c[i]=='G')
		{
			cur[s[i]-'a']++;
			for(int j=0;j<26;j++)can[i][j]=0;
			can[i][s[i]-'a']=1;
		}
		rep(i,len)
		{
			if (c[i]=='Y')
			{
				cur[s[i]-'a']++;
				can[i][s[i]-'a']=0;
			}
			else
			if (c[i]=='B')
			{
				r[s[i]-'a']=min(r[s[i]-'a'],cur[s[i]-'a']);
				can[i][s[i]-'a']=0;
			}
			l[s[i]-'a']=max(l[s[i]-'a'],cur[s[i]-'a']);
		}
	}

	nwf.S=0; nwf.T=len+26+1; nwf.clear();
	rep(i,len)nwf.addedge(nwf.S,i,1,0);
	rep(i,len)for(int j=0;j<26;j++)
		if (can[i][j])
		{
			//cout<<i<<" "<<j<<endl;
			id[i][j]=nwf.top+1;
			nwf.addedge(i,len+j+1,1,0);
		}
	//cout<<l[4]<<" "<<r[4]<<endl;
	for(int i=0;i<26;i++)
	{
		if (l[i]>r[i])continue;
		nwf.addedge(len+i+1,nwf.T,l[i],-100000);
		nwf.addedge(len+i+1,nwf.T,r[i]-l[i],0);
	}

	nwf.work();

	rep(i,len)for(char c='a';c<='z';c++)
		if (id[i][c-'a']!=0 && nwf.edge[id[i][c-'a']].w==0)putchar(c);
	puts("");
}