mplayer+科大讯飞离线语音包

企业开发 2022-05-29 14:09:19 阅读次数: 0

在这里插入图片描述

文章目录

【1】mplayer

1.将mplayer放到开发板的bin目录下

system(“mplayer 1.avi &”); // &为后台播放视频

播放音乐

system(“madplay -a -10 xxx.mp3”);//-a 音量控制 范围（-175~+18）

暂停播放

system(“killall -STOP madplay &”);

恢复播放

system(“killall -CONT madplay &”);

停止

system(“killall madplay”);

madplay终止

system(“killall -9 madplay”);

思路 ：如何保证将电脑的音频或视频传输给开发板，保持数据完整

1.先发送文件大小
2.再发送数据内容
3.检测发送的数据与接收的数据大小是否一致

【2】科大讯飞离线语音包

链接：https://pan.baidu.com/s/1Dgo8HGs3_Bfw05M0dCbS4A?pwd=vxpz
提取码：vxpz
README文件

README for Linux_aitalk

bin：
|-- msc
    |-- msc.cfg（作用：msc调试、生成msc日志）
	|-- res
		|-- asr（资源文件）
	|-- wav
		|-- ddhghlj.pcm
		|-- ddhgdw.pcm
	|-- call.bnf
		
doc：
|-- readme.txt

include：调用SDK所需头文件

libs：
|-- x86
	|-- libmsc.so（32位动态库）
|-- x64
	|-- libmsc.so（64位动态库）

samples：
|-- asr_sample
	|-- asr_sample.c
	|-- Makefile
	|-- 32bit_make.sh
	|-- 64bit_make.sh
|-- asr_record_sample
	|-- formats.h
	|-- linuxrec.h
	|-- linuxrec.c
	|-- speech_recognizer.h
	|-- speech_recognizer.c
	|-- asr_record_sample.c
	|-- Makefile
	|-- 32bit_make.sh
	|-- 64bit_make.sh

asr_offline_sample.c

#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>

#include "../../include/qisr.h"
#include "../../include/msp_cmn.h"
#include "../../include/msp_errors.h"

#define SAMPLE_RATE_16K     (16000)
#define SAMPLE_RATE_8K      (8000)
#define MAX_GRAMMARID_LEN   (32)
#define MAX_PARAMS_LEN      (1024)

const char * ASR_RES_PATH        = "fo|res/asr/common.jet"; //离线语法识别资源路径
const char * GRM_BUILD_PATH      = "res/asr/GrmBuilld"; //构建离线语法识别网络生成数据保存路径
const char * GRM_FILE            = "call.bnf"; //构建离线识别语法网络所用的语法文件
const char * LEX_NAME            = "contact"; //更新离线识别语法的contact槽（语法文件为此示例中使用的call.bnf）

typedef struct _UserData {
    
    
	int     build_fini; //标识语法构建是否完成
	int     update_fini; //标识更新词典是否完成
	int     errcode; //记录语法构建或更新词典回调错误码
	char    grammar_id[MAX_GRAMMARID_LEN]; //保存语法构建返回的语法ID
}UserData;


const char *get_audio_file(void); //选择进行离线语法识别的语音文件
int build_grammar(UserData *udata); //构建离线识别语法网络
int update_lexicon(UserData *udata); //更新离线识别语法词典
int run_asr(UserData *udata); //进行离线语法识别

const char* get_audio_file(void)
{
    
    
	char key = 0;
	while(key != 27) //按Esc则退出
	{
    
    
		printf("请选择音频文件：\n");
		printf("1.打电话给丁伟\n");
		printf("2.打电话给黄辣椒\n");
		key = getchar();
		getchar();
		switch(key)
		{
    
    
		case '1':
			printf("\n1.打电话给丁伟\n");
			return "wav/ddhgdw.pcm";
		case '2':
			printf("\n2.打电话给黄辣椒\n");
			return "wav/ddhghlj.pcm";
		default:
			continue;
		}
	}
	exit(0);
	return NULL;
}

int build_grm_cb(int ecode, const char *info, void *udata)
{
    
    
	UserData *grm_data = (UserData *)udata;

	if (NULL != grm_data) {
    
    
		grm_data->build_fini = 1;
		grm_data->errcode = ecode;
	}

	if (MSP_SUCCESS == ecode && NULL != info) {
    
    
		printf("构建语法成功！ 语法ID:%s\n", info);
		if (NULL != grm_data)
			snprintf(grm_data->grammar_id, MAX_GRAMMARID_LEN - 1, info);
	}
	else
		printf("构建语法失败！%d\n", ecode);

	return 0;
}

int build_grammar(UserData *udata)
{
    
    
	FILE *grm_file                           = NULL;
	char *grm_content                        = NULL;
	unsigned int grm_cnt_len                 = 0;
	char grm_build_params[MAX_PARAMS_LEN]    = {
    
    NULL};
	int ret                                  = 0;

	grm_file = fopen(GRM_FILE, "rb");	
	if(NULL == grm_file) {
    
    
		printf("打开\"%s\"文件失败！[%s]\n", GRM_FILE, strerror(errno));
		return -1; 
	}

	fseek(grm_file, 0, SEEK_END);
	grm_cnt_len = ftell(grm_file);
	fseek(grm_file, 0, SEEK_SET);

	grm_content = (char *)malloc(grm_cnt_len + 1);
	if (NULL == grm_content)
	{
    
    
		printf("内存分配失败!\n");
		fclose(grm_file);
		grm_file = NULL;
		return -1;
	}
	fread((void*)grm_content, 1, grm_cnt_len, grm_file);
	grm_content[grm_cnt_len] = '\0';
	fclose(grm_file);
	grm_file = NULL;

	snprintf(grm_build_params, MAX_PARAMS_LEN - 1, 
		"engine_type = local, \
		asr_res_path = %s, sample_rate = %d, \
		grm_build_path = %s, ",
		ASR_RES_PATH,
		SAMPLE_RATE_16K,
		GRM_BUILD_PATH
		);
	ret = QISRBuildGrammar("bnf", grm_content, grm_cnt_len, grm_build_params, build_grm_cb, udata);

	free(grm_content);
	grm_content = NULL;

	return ret;
}

int update_lex_cb(int ecode, const char *info, void *udata)
{
    
    
	UserData *lex_data = (UserData *)udata;

	if (NULL != lex_data) {
    
    
		lex_data->update_fini = 1;
		lex_data->errcode = ecode;
	}

	if (MSP_SUCCESS == ecode)
		printf("更新词典成功！\n");
	else
		printf("更新词典失败！%d\n", ecode);

	return 0;
}

int update_lexicon(UserData *udata)
{
    
    
	const char *lex_content                   = "丁伟\n黄辣椒";
	unsigned int lex_cnt_len                  = strlen(lex_content);
	char update_lex_params[MAX_PARAMS_LEN]    = {
    
    NULL}; 

	snprintf(update_lex_params, MAX_PARAMS_LEN - 1, 
		"engine_type = local, text_encoding = UTF-8, \
		asr_res_path = %s, sample_rate = %d, \
		grm_build_path = %s, grammar_list = %s, ",
		ASR_RES_PATH,
		SAMPLE_RATE_16K,
		GRM_BUILD_PATH,
		udata->grammar_id);
	return QISRUpdateLexicon(LEX_NAME, lex_content, lex_cnt_len, update_lex_params, update_lex_cb, udata);
}

int run_asr(UserData *udata)
{
    
    
	char asr_params[MAX_PARAMS_LEN]    = {
    
    NULL};
	const char *rec_rslt               = NULL;
	const char *session_id             = NULL;
	const char *asr_audiof             = NULL;
	FILE *f_pcm                        = NULL;
	char *pcm_data                     = NULL;
	long pcm_count                     = 0;
	long pcm_size                      = 0;
	int last_audio                     = 0;
	int aud_stat                       = MSP_AUDIO_SAMPLE_CONTINUE;
	int ep_status                      = MSP_EP_LOOKING_FOR_SPEECH;
	int rec_status                     = MSP_REC_STATUS_INCOMPLETE;
	int rss_status                     = MSP_REC_STATUS_INCOMPLETE;
	int errcode                        = -1;

	asr_audiof = get_audio_file();
	f_pcm = fopen(asr_audiof, "rb");
	if (NULL == f_pcm) {
    
    
		printf("打开\"%s\"失败！[%s]\n", f_pcm, strerror(errno));
		goto run_error;
	}
	fseek(f_pcm, 0, SEEK_END);
	pcm_size = ftell(f_pcm);
	fseek(f_pcm, 0, SEEK_SET);
	pcm_data = (char *)malloc(pcm_size);
	if (NULL == pcm_data)
		goto run_error;
	fread((void *)pcm_data, pcm_size, 1, f_pcm);
	fclose(f_pcm);
	f_pcm = NULL;

	//离线语法识别参数设置
	snprintf(asr_params, MAX_PARAMS_LEN - 1, 
		"engine_type = local, \
		asr_res_path = %s, sample_rate = %d, \
		grm_build_path = %s, local_grammar = %s, \
		result_type = xml, result_encoding = UTF-8, ",
		ASR_RES_PATH,
		SAMPLE_RATE_16K,
		GRM_BUILD_PATH,
		udata->grammar_id
		);
	session_id = QISRSessionBegin(NULL, asr_params, &errcode);
	if (NULL == session_id)
		goto run_error;
	printf("开始识别...\n");

	while (1) {
    
    
		unsigned int len = 6400;

		if (pcm_size < 12800) {
    
    
			len = pcm_size;
			last_audio = 1;
		}

		aud_stat = MSP_AUDIO_SAMPLE_CONTINUE;

		if (0 == pcm_count)
			aud_stat = MSP_AUDIO_SAMPLE_FIRST;

		if (len <= 0)
			break;

		printf(">");
		fflush(stdout);
		errcode = QISRAudioWrite(session_id, (const void *)&pcm_data[pcm_count], len, aud_stat, &ep_status, &rec_status);
		if (MSP_SUCCESS != errcode)
			goto run_error;

		pcm_count += (long)len;
		pcm_size -= (long)len;

		//检测到音频结束
		if (MSP_EP_AFTER_SPEECH == ep_status)
			break;

		usleep(150 * 1000); //模拟人说话时间间隙
	}
	//主动点击音频结束
	QISRAudioWrite(session_id, (const void *)NULL, 0, MSP_AUDIO_SAMPLE_LAST, &ep_status, &rec_status);

	free(pcm_data);
	pcm_data = NULL;

	//获取识别结果
	while (MSP_REC_STATUS_COMPLETE != rss_status && MSP_SUCCESS == errcode) {
    
    
		rec_rslt = QISRGetResult(session_id, &rss_status, 0, &errcode);
		usleep(150 * 1000);
	}
	printf("\n识别结束：\n");
	printf("=============================================================\n");
	if (NULL != rec_rslt)
		printf("%s\n", rec_rslt);
	else
		printf("没有识别结果！\n");
	printf("=============================================================\n");

	goto run_exit;

run_error:
	if (NULL != pcm_data) {
    
    
		free(pcm_data);
		pcm_data = NULL;
	}
	if (NULL != f_pcm) {
    
    
		fclose(f_pcm);
		f_pcm = NULL;
	}
run_exit:
	QISRSessionEnd(session_id, NULL);
	return errcode;
}

int main(int argc, char* argv[])
{
    
    
	const char *login_config    = "appid = a54c3878"; //登录参数
	UserData asr_data; 
	int ret                     = 0 ;
	char c;

	ret = MSPLogin(NULL, NULL, login_config); //第一个参数为用户名，第二个参数为密码，传NULL即可，第三个参数是登录参数
	if (MSP_SUCCESS != ret) {
    
    
		printf("登录失败：%d\n", ret);
		goto exit;
	}

	memset(&asr_data, 0, sizeof(UserData));
	printf("构建离线识别语法网络...\n");
	ret = build_grammar(&asr_data);  //第一次使用某语法进行识别，需要先构建语法网络，获取语法ID，之后使用此语法进行识别，无需再次构建
	if (MSP_SUCCESS != ret) {
    
    
		printf("构建语法调用失败！\n");
		goto exit;
	}
	while (1 != asr_data.build_fini)
		usleep(300 * 1000);
	if (MSP_SUCCESS != asr_data.errcode)
		goto exit;
	printf("离线识别语法网络构建完成，开始识别...\n");	
	ret = run_asr(&asr_data);
	if (MSP_SUCCESS != ret) {
    
    
		printf("离线语法识别出错: %d \n", ret);
		goto exit;
	}

	printf("更新离线语法词典...\n");
	ret = update_lexicon(&asr_data);  //当语法词典槽中的词条需要更新时，调用QISRUpdateLexicon接口完成更新
	if (MSP_SUCCESS != ret) {
    
    
		printf("更新词典调用失败！\n");
		goto exit;
	}
	while (1 != asr_data.update_fini)
		usleep(300 * 1000);
	if (MSP_SUCCESS != asr_data.errcode)
		goto exit;
	printf("更新离线语法词典完成，开始识别...\n");
	ret = run_asr(&asr_data);
	if (MSP_SUCCESS != ret) {
    
    
		printf("离线语法识别出错: %d \n", ret);
		goto exit;
	}

exit:
	MSPLogout();
	printf("请按任意键退出...\n");
	getchar();
	return 0;
}

Makefile

#common makefile header

DIR_INC = ../../include
DIR_BIN = ../../bin
DIR_LIB = ../../libs

TARGET	= asr_offline_sample
BIN_TARGET = $(DIR_BIN)/$(TARGET)

CROSS_COMPILE = 
CFLAGS = -g -Wall -I$(DIR_INC)

ifdef LINUX64
LDFLAGS := -L$(DIR_LIB)/x64
else
LDFLAGS := -L$(DIR_LIB)/x86 
endif
LDFLAGS += -lmsc -lrt -ldl -lpthread  -lstdc++

OBJECTS := $(patsubst %.c,%.o,$(wildcard *.c))

$(BIN_TARGET) : $(OBJECTS)
	$(CROSS_COMPILE)gcc $(CFLAGS) $^ -o $@ $(LDFLAGS)

%.o : %.c
	$(CROSS_COMPILE)gcc -c $(CFLAGS) $< -o $@
clean:
	@rm -f *.o $(BIN_TARGET)

.PHONY:clean

#common makefile foot

在这里插入图片描述
call.bnf

#BNF+IAT 1.0 UTF-8;
!grammar call;
!slot <want>;
!slot <dialpre>;
!slot <dialsuf>;
!slot <contact>;

!start <callstart>;
<callstart>:[<want>]<dial>;
<want>:我想|我要|请|帮我|我想要|请帮我;
<dial>:<dialpre><contact>[<dialsuf>];
<dialpre>:打电话给!id(10001)|打给!id(10001)|拨打!id(10001)|拨打电话给!id(10001)|呼叫!id(10001)|
打一个电话给!id(10001)|打个电话给!id(10001)|给|拨通!id(10001)|
接通!id(10001)|呼叫!id(10001)|呼叫给!id(10001)|打!id(10001);
<dialsuf>:打电话!id(10001)|打个电话!id(10001)|打一个电话!id(10001)|
拨打电话!id(10001)|拨电话!id(10001)|拨个电话!id(10001)|呼个电话!id(10001)|
的电话!id(10001)|的号码!id(10001)|的手机!id(10001)|
的办公电话!id(10001)|的移动号码!id(10001)|的联通号码!id(10001)|
的电信号码!id(10001)|客服电话!id(10001);
<contact>:丁伟;

猜你喜欢

转载自blog.csdn.net/m0_45463480/article/details/125033058
今日推荐
Apache Doris 2.0.10 版本正式发布！
开源日报 | 大模型开战；大模型独角兽被曝卖身；周鸿祎建议谷歌开源所有产品；最大开源AI社区提供1000万美元共享GPU
开源日报 | Chrome内置Gemini的意义不在于Gemini；中国AI追随之路的五大误区；ECharts创始人“下海”养鱼；谷歌I/O开发者大会什么都有，只是没有惊喜
微软回应中国区AI团队“打包赴美”传闻
基于大语言模型的开源知识库问答系统 MaxKB GitHub Star 数量突破 5,000 个！
美国拟限制 AI 大模型出口中国和俄罗斯
苹果将与 OpenAI 达成协议，将 ChatGPT 应用于 iPhone
周排行
阿里云短信服务平台注册
Windows下的字符串处理(1)
sqoop: mysql导入数据到hdfs, hive, hbase
commons.lang中常用的工具类
离线安装PostgreSQL11.6
使用PyTorch简单实现卷积神经网络模型
一文彻底搞定谱聚类
一道面试题引发的血案
One Chat for Mac(聊天工具)
TCP/IP的底层队列是如何实现的？
每日归档
更多
2024-05-17(34)
2024-05-16(6)
2024-05-15(24)
2024-05-14(0)
2024-05-13(18)
2024-05-12(0)
2024-05-11(38)
2024-05-10(38)
2024-05-09(35)
2024-05-08(42)

代码天地 © 2018 codetd.com

境外服务器   最新电视剧2022