语音识别的大规模汉语树形词典搜索速度快如闪电

Posted 2021-01-28 wicnwicnwh

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了语音识别的大规模汉语树形词典搜索速度快如闪电相关的知识，希望对你有一定的参考价值。

#include "stdafx.h"
#include "YuyinTree.h"
#include "YuyinTreeDlg.h"
#ifdef _DEBUG
#define new DEBUG_NEW
#undef THIS_FILE
static char THIS_FILE[] = __FILE__;
#endif
/////////////////////////////////////////////////////////////////////////////
// CAboutDlg dialog used for App About
CStoredPinyin* m_storedPinyin[30000];
CWordTree* curTree=new CWordTree; //指向词语树指针
CString MaxLenghci=‘/‘‘; //存储该单词包括的最大子词
CString Tempci; //临时词
int StoredNum=0; //已存入查询数组中拼音个数
class CAboutDlg : public CDialog
{
public:
CAboutDlg();
// Dialog Data
//{{AFX_DATA(CAboutDlg)
enum { IDD = IDD_ABOUTBOX };
//}}AFX_DATA
// ClassWizard generated virtual function overrides
//{{AFX_VIRTUAL(CAboutDlg)
protected:
virtual void DoDataExchange(CDataExchange* pDX); // DDX/DDV support
//}}AFX_VIRTUAL
// Implementation
protected:
//{{AFX_MSG(CAboutDlg)
//}}AFX_MSG
DECLARE_MESSAGE_MAP()
};
CAboutDlg::CAboutDlg() : CDialog(CAboutDlg::IDD)
{
//{{AFX_DATA_INIT(CAboutDlg)
//}}AFX_DATA_INIT
}
void CAboutDlg::DoDataExchange(CDataExchange* pDX)
{
CDialog::DoDataExchange(pDX);
//{{AFX_DATA_MAP(CAboutDlg)
//}}AFX_DATA_MAP
}
BEGIN_MESSAGE_MAP(CAboutDlg, CDialog)
//{{AFX_MSG_MAP(CAboutDlg)
// No message handlers
//}}AFX_MSG_MAP
END_MESSAGE_MAP()
/////////////////////////////////////////////////////////////////////////////
// CYuyinTreeDlg dialog
CYuyinTreeDlg::CYuyinTreeDlg(CWnd* pParent /*=NULL*/)
: CDialog(CYuyinTreeDlg::IDD, pParent)
{
//{{AFX_DATA_INIT(CYuyinTreeDlg)
m_inputTongyin = _T("");
//}}AFX_DATA_INIT
// Note that LoadIcon does not require a subsequent DestroyIcon in Win32
m_hIcon = AfxGetApp()->LoadIcon(IDR_MAINFRAME);
}
void CYuyinTreeDlg::DoDataExchange(CDataExchange* pDX)
{
CDialog::DoDataExchange(pDX);
//{{AFX_DATA_MAP(CYuyinTreeDlg)
DDX_Text(pDX, IDC_PINYIN_EQUAL, m_inputTongyin);
//}}AFX_DATA_MAP
}
BEGIN_MESSAGE_MAP(CYuyinTreeDlg, CDialog)
//{{AFX_MSG_MAP(CYuyinTreeDlg)
ON_WM_SYSCOMMAND()
ON_WM_PAINT()
ON_WM_QUERYDRAGICON()
ON_BN_CLICKED(IDC_BTN_READFILE, OnBtnReadfile)
ON_BN_CLICKED(IDC_BTN_TEST_PANDCHI, OnBtnTestPandchi)
ON_BN_CLICKED(IDC_BTN_TEST_TONGYINCI, OnBtnTestTongyinci)
ON_BN_CLICKED(IDC_YUYIN_QUERY, OnTongyinQuery)
ON_NOTIFY(TVN_SELCHANGED, IDC_YUYIN_TREE, OnSelchangedYuyinTree)
//}}AFX_MSG_MAP
END_MESSAGE_MAP()
/////////////////////////////////////////////////////////////////////////////
// CYuyinTreeDlg message handlers
BOOL CYuyinTreeDlg::OnInitDialog()
{
CDialog::OnInitDialog();
// Add "About..." menu item to system menu.
// IDM_ABOUTBOX must be in the system command range.
ASSERT((IDM_ABOUTBOX & 0xFFF0) == IDM_ABOUTBOX);
ASSERT(IDM_ABOUTBOX < 0xF000);
CMenu* pSysMenu = GetSystemMenu(FALSE);
if (pSysMenu != NULL)
{
CString strAboutMenu;
strAboutMenu.LoadString(IDS_ABOUTBOX);
if (!strAboutMenu.IsEmpty())
{
pSysMenu->AppendMenu(MF_SEPARATOR);
pSysMenu->AppendMenu(MF_STRING, IDM_ABOUTBOX, strAboutMenu);
}
}
// Set the icon for this dialog. The framework does this automatically
// when the application‘s main window is not a dialog
SetIcon(m_hIcon, TRUE); // Set big icon
SetIcon(m_hIcon, FALSE); // Set small icon
// TODO: Add extra initialization here
//程序添加的代码:
//初始化已读单词数组
for(int i=0;i<30000;i++)
{
m_storedPinyin[i]=NULL;
}
FILE *InputFile=NULL; //读取文件指针
CTreeNode* CurNode=NULL; //当前节点
CTreeNode* FatherNode=NULL; //当前节点的父节点
TCHAR NodeItem[100]; //节点的全部数据
TCHAR NodePinyin[51]; //节点的拼音
TCHAR NodeHanzi[50]; //节点汉字
int ZiMuNum,j,k;
int WordNum=0; //统计词的数目
BOOL IsFirst=TRUE; //标志是否是根节点的第一个孩子节点
CTreeNode* TempfatherNode=NULL; //临时父节点
CTreeNode* TempfatherNode1=NULL; //临时父节点
InputFile=fopen("LexiconTree.txt","r");
while(!feof(InputFile))
{
WordNum++;
fscanf(InputFile,"%s",NodeItem);
ZiMuNum=0; //ZiMuNum为读取字母个数
while(NodeItem[ZiMuNum]!=‘,‘)//读取汉字
{
NodeHanzi[ZiMuNum]=NodeItem[ZiMuNum];
ZiMuNum++;
}
NodeHanzi[ZiMuNum]=‘/0‘;// 此时NodeHanzi存储了汉字
ZiMuNum++;
while(NodeItem[ZiMuNum]!=‘,‘)//读取词号
{
ZiMuNum++;
}
ZiMuNum++;
j=ZiMuNum;
k=0;
while(NodeItem[j]!=‘/0‘) //读拼音
{
NodePinyin[k]=NodeItem[j];
k++;
j++;
}
NodePinyin[k-1]=‘/0‘; //此时NodePinyin存储了拼音
CString TempString; //暂时存储拼音
TempString=NodePinyin;
CurNode=new CTreeNode;
CurNode->m_Pinyin=TempString;
CurNode->m_Word=NodeHanzi;
CurNode->m_pParent=NULL;
CurNode->m_pchild=NULL;
CurNode->m_pneighbour=NULL;
CurNode->m_pchar=NULL;
CurNode->m_CurNum=0;
BOOL Fviewed=FALSE; //是否查到的标志,默认为没查到
BOOL FParent=FALSE; //接点是否有父节点的标志，如a,;a,ba,;a,ba,hao,
BOOL FEqual=FALSE; //相同发音词标志，如a,啊，阿，呵
for(int temp=0;temp<30000;temp++) //temp为迭代次数,先检查看看是否已经插入了词
{
if(m_storedPinyin[temp]!=NULL) //看当前查看的数据项是否有值,若有进入循环,否则跳过
{
int t=-1;
t=TempString.Find(m_storedPinyin[temp]->m_StoredPinyin); //find 的意思是找到整个匹配串
if(t>=0) //若有重叠项,进入语句,插数据项
{
if(TempString==m_storedPinyin[temp]->m_StoredPinyin) //若两者拼音完全相同,把汉字插入树节点项的相似字数组中
{
Fviewed=TRUE; //查到
FEqual=TRUE; //相同词标记TRUE
int TempCurNum=(m_storedPinyin[temp]->m_pcurPosition)->m_CurNum;
CTongyinci* temp1=new CTongyinci;
temp1->next=NULL;
temp1->m_data=new char[50];
strcpy(temp1->m_data,NodeHanzi);
CTongyinci* temp2=(m_storedPinyin[temp]->m_pcurPosition)->m_pchar;
CTongyinci* pre;
while(temp2!=NULL)
{
pre=temp2;
temp2=temp2->next;
}
pre->next=temp1;
(m_storedPinyin[temp]->m_pcurPosition)->m_CurNum=(m_storedPinyin[temp]->m_pcurPosition)->m_CurNum+1;
break;
}
else if(t==0) //否则,插入树节点项的子节点中,必需保证从第一个字对齐并且在待检查接点中重叠部分后一位为",",检查
{ //比如a和an不能是父子节点关系;la和a不能是父子节点关系
Tempci=m_storedPinyin[temp]->m_StoredPinyin;
int Strsize=Tempci.GetLength(); //问题
if(TempString.GetAt(Strsize)==‘,‘) //若满足重叠部分后一位为","
{
FParent=TRUE; //有父亲节点
if(Strsize>=MaxLenghci.GetLength())
{
MaxLenghci=Tempci; //注意
TempfatherNode1=m_storedPinyin[temp]->m_pcurPosition; //注意
}
Fviewed=TRUE; //查到
}
}
}
}
else
continue;
}
if(FParent&&!FEqual)
{
CurNode->m_CurNum++; //当前同音字数加1
/* char** pTempChar=new char*[50]; //开辟一个而维数组，临时值向同音字的指针
for(int l=0;l<50;l++)
{
pTempChar[l]=new char[50];
pTempChar[l][0]=‘/0‘;
}
*/
CTongyinci* temp=new CTongyinci;
temp->next=NULL;
temp->m_data=new char[50];
strcpy(temp->m_data,NodeHanzi);
CurNode->m_pchar=temp;
curTree->InsertNode(&CurNode,&TempfatherNode1); //在查到的节点处插入子节点
CStoredPinyin* TempStorPinyin1=new CStoredPinyin; //创建临时待查询对象
TempStorPinyin1->m_pcurPosition=CurNode;
TempStorPinyin1->m_StoredPinyin=TempString;
TempStorPinyin1->m_pchar=temp;
m_storedPinyin[StoredNum++]=TempStorPinyin1; //将产生节点存放已访问数组中
}
MaxLenghci=‘/0‘; //恢复用到的两个变量到初值
TempfatherNode1=NULL;
if(!Fviewed) //如果在已存词表中找不到该词则在根节点插入新词,并存放在已访问数组中
{
if(IsFirst==TRUE)
{
TempfatherNode=curTree->m_Root;
}
CurNode->m_CurNum++; //当前同音字数加1
/* char** pTempChar=new char*[50]; //开辟一个而维数组，临时值向同音字的指针,存100个词
for(int l=0;l<50;l++)
{
pTempChar[l]=new char[10];
pTempChar[l][0]=‘/0‘;
}
CurNode->m_pchar=pTempChar;
strcpy(CurNode->m_pchar[0],NodeHanzi);*/
CTongyinci* temp=new CTongyinci;
temp->next=NULL;
temp->m_data=new char[50];
strcpy(temp->m_data,NodeHanzi);
CurNode->m_pchar=temp;
TempfatherNode=curTree->InsertFirstNode(&CurNode,&TempfatherNode); //在根节点插入新词
IsFirst=FALSE; //以后插入的点全不是第一个节点
CStoredPinyin* TempStorPinyin; //创建临时待查询对象
TempStorPinyin=new CStoredPinyin;
TempStorPinyin->m_pcurPosition=CurNode;
TempStorPinyin->m_StoredPinyin=NodePinyin;
TempStorPinyin->m_pchar=temp;
//排错flag pass 10月21号
m_storedPinyin[StoredNum]=TempStorPinyin; //将产生节点存放已访问数组中
StoredNum++;
}
}
char buffer[50];
sprintf(buffer,"已读取单词: %d",WordNum);
AfxMessageBox(buffer);
BrowseYuyin();
return TRUE; // return TRUE unless you set the focus to a control
}
void CYuyinTreeDlg::OnSysCommand(UINT nID, LPARAM lParam)
{
if ((nID & 0xFFF0) == IDM_ABOUTBOX)
{
CAboutDlg dlgAbout;
dlgAbout.DoModal();
}
else
{
CDialog::OnSysCommand(nID, lParam);
}
}
// If you add a minimize button to your dialog, you will need the code below
// to draw the icon. For MFC applications using the document/view model,
// this is automatically done for you by the framework.
void CYuyinTreeDlg::OnPaint()
{
if (IsIconic())
{
CPaintDC dc(this); // device context for painting
SendMessage(WM_ICONERASEBKGND, (WPARAM) dc.GetSafeHdc(), 0);
// Center icon in client rectangle
int cxIcon = GetSystemMetrics(SM_CXICON);
int cyIcon = GetSystemMetrics(SM_CYICON);
CRect rect;
GetClientRect(&rect);
int x = (rect.Width() - cxIcon + 1) / 2;
int y = (rect.Height() - cyIcon + 1) / 2;
// Draw the icon
dc.DrawIcon(x, y, m_hIcon);
}
else
{
CDialog::OnPaint();
}
}
// The system calls this to obtain the cursor to display while the user drags
// the minimized window.
HCURSOR CYuyinTreeDlg::OnQueryDragIcon()
{
return (HCURSOR) m_hIcon;
}
// 添加的主要代码
void CYuyinTreeDlg::OnBtnReadfile() //测试树第一层节点
{
//检验第一层节点插入是否正确，结果正确
CTreeNode* root=curTree->m_Root;
CTreeNode* Temp=root->m_pchild;
while(Temp!=NULL)
{
AfxMessageBox(Temp->m_Pinyin);
Temp=Temp->m_pneighbour;
}
}
CWordTree::CWordTree() //树的初始化
{
m_Root=new CTreeNode;
m_Root->m_Pinyin="";
m_Root->m_pParent=NULL;
m_Root->m_pchild=NULL;
m_Root->m_pneighbour=NULL;
m_Root->m_CurNum=0;
m_Root->m_Word="";
m_Root->m_pchar=NULL;
}
void CWordTree::CreateTree()
{
}
CTreeNode* CWordTree::InsertFirstNode(CTreeNode** newNode,CTreeNode** parent) //插入第一层节点
{
CTreeNode* WilladNode=*newNode;
CTreeNode* FatherNode=*parent;
if(FatherNode==curTree->m_Root)
{
FatherNode->m_pchild=WilladNode;
WilladNode->m_pParent=curTree->m_Root;
}
else
{
FatherNode->m_pneighbour=WilladNode;
WilladNode->m_pParent=curTree->m_Root;
}
return WilladNode;
}
void CWordTree::InsertNode(CTreeNode** newNode,CTreeNode** parent) //插入某层节点的孩子节点
{
CTreeNode* WilladNode=*newNode;
CTreeNode* FatherNode=*parent;
CTreeNode* Temp=NULL;
if(FatherNode->m_pchild==NULL)
{
FatherNode->m_pchild=WilladNode;
WilladNode->m_pParent=FatherNode;
}
else
{
Temp=FatherNode->m_pchild;
while(Temp->m_pneighbour!=NULL)
{
Temp=Temp->m_pneighbour;
}
Temp->m_pneighbour=WilladNode;
WilladNode->m_pParent=FatherNode;
}
}
void CYuyinTreeDlg::OnBtnTestPandchi() //测试树的相邻层接点
{
CTreeNode* root=curTree->m_Root;
CTreeNode* Temp=root->m_pchild;
AfxMessageBox(Temp->m_Pinyin);
Temp=Temp->m_pchild;
while(Temp!=NULL)
{
AfxMessageBox(Temp->m_Pinyin);
Temp=Temp->m_pneighbour;
}
}
void CYuyinTreeDlg::OnBtnTestTongyinci() //测试发音为‘a‘的同音词
{
CTreeNode* NodeA=curTree->m_Root->m_pchild;
CTongyinci* temp=NodeA->m_pchar;
while(temp!=NULL)
{
AfxMessageBox(temp->m_data);
temp=temp->next;
}
}
void CYuyinTreeDlg::BrowseYuyin() //遍历4层语音树
{
CTreeCtrl* pCtrl = (CTreeCtrl*) GetDlgItem(IDC_YUYIN_TREE);
ASSERT(pCtrl != NULL);
TVINSERTSTRUCT tvInsert;
tvInsert.hParent = NULL;
tvInsert.hInsertAfter = NULL;
tvInsert.item.mask = TVIF_TEXT;
tvInsert.item.pszText = _T("词拼音");
HTREEITEM hRoot = pCtrl->InsertItem(&tvInsert);
CTreeNode* root=curTree->m_Root;
CTreeNode* Temp=root->m_pchild;
HTREEITEM hPA=NULL;
HTREEITEM hPB=NULL;
HTREEITEM hPC=NULL;
while(Temp!=NULL)
{
hPA=pCtrl->InsertItem(TVIF_TEXT,
_T(Temp->m_Pinyin), 0, 0, 0, 0, 0, hRoot, NULL);
CTreeNode* TempNode1=Temp->m_pchild;
while(TempNode1!=NULL)
{
hPB=pCtrl->InsertItem(TVIF_TEXT,
_T(TempNode1->m_Pinyin), 0, 0, 0, 0, 0, hPA, NULL);
CTreeNode* TempNode2=TempNode1->m_pchild;
while(TempNode2!=NULL)
{
hPC=pCtrl->InsertItem(TVIF_TEXT,
_T(TempNode2->m_Pinyin), 0, 0, 0, 0, 0, hPB, NULL);
CTreeNode* TempNode3=TempNode2->m_pchild;
while(TempNode3!=NULL)
{
pCtrl->InsertItem(TVIF_TEXT,
_T(TempNode3->m_Pinyin), 0, 0, 0, 0, 0, hPC, NULL);
TempNode3=TempNode3->m_pneighbour;
}
TempNode2=TempNode2->m_pneighbour;
}
TempNode1=TempNode1->m_pneighbour;
}
Temp=Temp->m_pneighbour;
}
}
void CYuyinTreeDlg::OnTongyinQuery() //查询拼音的同音字,放入Combox中
{
// TODO: Add your control notification handler code here
UpdateData(TRUE);
CString StringWilqury=m_inputTongyin;
CString Tempci;
BOOL FExit=FALSE; //该拼音在树中是否存在标志
int TempNum;
CComboBox* tongyinc=(CComboBox*)GetDlgItem(IDC_PINYIN_RESULT);
ASSERT(tongyinc != NULL);
tongyinc->ResetContent();
ASSERT(tongyinc->GetCount() == 0);
for(int temp=0;temp<30000;temp++)
{
if(m_storedPinyin[temp]!=NULL&&m_storedPinyin[temp]->m_StoredPinyin==StringWilqury)
{
TempNum=m_storedPinyin[temp]->m_pcurPosition->m_CurNum;
// for(int i=0;i<TEMPNUM;I++) Tempci="m_storedPinyin[temp]-" {>m_pcurPosition->m_pchar[i];
// }
CTongyinci* temp1=m_storedPinyin[temp]->m_pcurPosition->m_pchar;
while(temp1!=NULL)
{
Tempci=temp1->m_data;
temp1=temp1->next;
tongyinc->AddString(Tempci);
}
tongyinc->SetWindowText(m_storedPinyin[temp]->m_pcurPosition->m_pchar->m_data);
FExit=TRUE;
}
}
if(!FExit)
tongyinc->SetWindowText("该拼音在树中不存在!");
}
void CYuyinTreeDlg::OnSelchangedYuyinTree(NMHDR* pNMHDR, LRESULT* pResult) //当单击鼠标时响应函数
{
NM_TREEVIEW* pNMTreeView = (NM_TREEVIEW*)pNMHDR;
// TODO: Add your control notification handler code here
TV_ITEM item=pNMTreeView->itemNew;
HTREEITEM hItem=item.hItem;
CTreeCtrl* pCtrl = (CTreeCtrl*) GetDlgItem(IDC_YUYIN_TREE);
ASSERT(pCtrl != NULL);
CString Tempci=pCtrl->GetItemText(hItem);
if(Tempci=="词拼音")
return;
BOOL FExit=FALSE; //该拼音在树中是否存在标志
int TempNum;
CComboBox* tongyinc=(CComboBox*)GetDlgItem(IDC_PINYIN_RESULT);
ASSERT(tongyinc != NULL);
tongyinc->ResetContent();
ASSERT(tongyinc->GetCount() == 0);
for(int temp=0;temp<30000;temp++)
{
if(m_storedPinyin[temp]!=NULL&&m_storedPinyin[temp]->m_StoredPinyin==Tempci)
{
TempNum=m_storedPinyin[temp]->m_pcurPosition->m_CurNum;
/* for(int i=0;i<TEMPNUM;I++) Tempci="m_storedPinyin[temp]-" {>m_pcurPosition->m_pchar[i];
tongyinc->AddString(Tempci);
}
*/
CTongyinci* temp1=m_storedPinyin[temp]->m_pcurPosition->m_pchar;
while(temp1!=NULL)
{
Tempci=temp1->m_data;
temp1=temp1->next;
tongyinc->AddString(Tempci);
}
tongyinc->SetWindowText(m_storedPinyin[temp]->m_pcurPosition->m_pchar->m_data);
FExit=TRUE;
}
}
if(!FExit)
tongyinc->SetWindowText("该拼音在树中不存在!");
*pResult = 0;
}

再分享一下我老师大神的人工智能教程吧。零基础！通俗易懂！风趣幽默！希望你也加入到我们人工智能的队伍中来！http://www.captainbed.net

以上是关于语音识别的大规模汉语树形词典搜索速度快如闪电的主要内容，如果未能解决你的问题，请参考以下文章

12-11锦上添花Flutter AI只能语音搜索功能实现

01 介绍

求助，语音识别的软件，口述录音能够识别转化成文字的

如何解释语音识别的技术原理？

语音中的关于语音识别的一些知识

关于语音识别的 Android 附加功能不起作用

语音识别的大规模汉语树形词典 搜索速度快如闪电

语音识别的大规模汉语树形词典搜索速度快如闪电