ROS实战( 三 )利用科大讯飞tts实现ROS下语音合成播报

一.前言

继上篇博客的内容,下面主要介绍流程:

我们从图中可以看出,首先xf_tts节点订阅了/voice/xf_tts_topic这个话题,这个话题的类型是std_msgs/String,然后通过调用科大讯飞在线合成代码形成节点将收到的文本输入进语音合成文件,文件类型是.wav,最后通过system函数来调用play命令,来播放.wav文件.

二.操作流程

首先默认你安装了ros,,并配置好了相关的路径和环境,没有安装的参考我这篇博客
https://blog.csdn.net/weixin_40522162/article/details/79244089
打开终端,cd到ros工作路径下源文件下,即/catkin_ws/src下,首先创建一个自己的包

catkin_create_pkg voice_system roscpp rospy std_msgs
cd到你下载的科大讯飞sdk下,即/samples下
cp tts_sample.c ~/catkin_ws/src/voice_system/src/
cd ../../include
cp * ~/catkin_ws/src/voice_system/include/
cd ~/catkin_ws/src/voice_system/src
mv tts_sample.c xf_tts.cpp
vim xf_tts.cpp

未修改的tts_sample.c如下:

/*
* 语音合成（Text To Speech，TTS）技术能够自动将任意文字实时转换为连续的
* 自然语音，是一种能够在任何时间、任何地点，向任何人提供语音信息服务的
* 高效便捷手段，非常符合信息时代海量数据、动态更新和个性化查询的需求。
*/

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

#include "qtts.h"
#include "msp_cmn.h"
#include "msp_errors.h"

/* wav音频头部格式 */
typedef struct _wave_pcm_hdr
{
    char            riff[4];                // = "RIFF"
    int     size_8;                 // = FileSize - 8
    char            wave[4];                // = "WAVE"
    char            fmt[4];                 // = "fmt "
    int     fmt_size;       // = 下一个结构体的大小 : 16

    short int       format_tag;             // = PCM : 1
    short int       channels;               // = 通道数 : 1
    int     samples_per_sec;        // = 采样率 : 8000 | 6000 | 11025 | 16000
    int     avg_bytes_per_sec;      // = 每秒字节数 : samples_per_sec * bits_per_sample / 8
    short int       block_align;            // = 每采样点字节数 : wBitsPerSample / 8
    short int       bits_per_sample;        // = 量化比特数: 8 | 16

    char            data[4];                // = "data";
    int     data_size;              // = 纯数据长度 : FileSize - 44 
} wave_pcm_hdr;

/* 默认wav音频头部数据 */
wave_pcm_hdr default_wav_hdr = 
{
    { 'R', 'I', 'F', 'F' },
    0,
    {'W', 'A', 'V', 'E'},
    {'f', 'm', 't', ' '},
    16,
    1,
    1,
    16000,
    32000,
    2,
    16,
    {'d', 'a', 't', 'a'},
    0  
};
/* 文本合成 */
int text_to_speech(const char* src_text, const char* des_path, const char* params)
{
    int          ret          = -1;
    FILE*        fp           = NULL;
    const char*  sessionID    = NULL;
    unsigned int audio_len    = 0;
    wave_pcm_hdr wav_hdr      = default_wav_hdr;
    int          synth_status = MSP_TTS_FLAG_STILL_HAVE_DATA;

    if (NULL == src_text || NULL == des_path)
    {
        printf("params is error!\n");
        return ret;
    }
    fp = fopen(des_path, "wb");
    if (NULL == fp)
    {
        printf("open %s error.\n", des_path);
        return ret;
    }
    /* 开始合成 */
    sessionID = QTTSSessionBegin(params, &ret);
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSSessionBegin failed, error code: %d.\n", ret);
        fclose(fp);
        return ret;
    }
    ret = QTTSTextPut(sessionID, src_text, (unsigned int)strlen(src_text), NULL);
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSTextPut failed, error code: %d.\n",ret);
        QTTSSessionEnd(sessionID, "TextPutError");
        fclose(fp);
        return ret;
    }
    printf("正在合成 ...\n");
    fwrite(&wav_hdr, sizeof(wav_hdr) ,1, fp); //添加wav音频头，使用采样率为16000
    while (1) 
    {
        /* 获取合成音频 */
        const void* data = QTTSAudioGet(sessionID, &audio_len, &synth_status, &ret);
        if (MSP_SUCCESS != ret)
            break;
        if (NULL != data)
        {
            fwrite(data, audio_len, 1, fp);
            wav_hdr.data_size += audio_len; //计算data_size大小
        }
        if (MSP_TTS_FLAG_DATA_END == synth_status)
            break;
        printf(">");
        usleep(150*1000); //防止频繁占用CPU
    }
    printf("\n");
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSAudioGet failed, error code: %d.\n",ret);
        QTTSSessionEnd(sessionID, "AudioGetError");
        fclose(fp);
        return ret;
    }
    /* 修正wav文件头数据的大小 */
    wav_hdr.size_8 += wav_hdr.data_size + (sizeof(wav_hdr) - 8);

    /* 将修正过的数据写回文件头部,音频文件为wav格式 */
    fseek(fp, 4, 0);
    fwrite(&wav_hdr.size_8,sizeof(wav_hdr.size_8), 1, fp); //写入size_8的值
    fseek(fp, 40, 0); //将文件指针偏移到存储data_size值的位置
    fwrite(&wav_hdr.data_size,sizeof(wav_hdr.data_size), 1, fp); //写入data_size的值
    fclose(fp);
    fp = NULL;
    /* 合成完毕 */
    ret = QTTSSessionEnd(sessionID, "Normal");
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSSessionEnd failed, error code: %d.\n",ret);
    }

    return ret;
}

int main(int argc, char* argv[])
{
    int         ret                  = MSP_SUCCESS;
    const char* login_params         = "appid = 5afcee34, work_dir = .";//登录参数,appid与msc库绑定,请勿随意改动
    /*
    * rdn:           合成音频数字发音方式
    * volume:        合成音频的音量
    * pitch:         合成音频的音调
    * speed:         合成音频对应的语速
    * voice_name:    合成发音人
    * sample_rate:   合成音频采样率
    * text_encoding: 合成文本编码格式
    *
    */
    const char* session_begin_params = "voice_name = xiaoyan, text_encoding = utf8, sample_rate = 16000, speed = 50, volume = 50, pitch = 50, rdn = 2";
    const char* filename             = "tts_sample.wav"; //合成的语音文件名称
    const char* text                 = "亲爱的用户，您好，这是一个语音合成示例，感谢您对科大讯飞语音技术的支持！科大讯飞是亚太地区最大的语音上市公司，股票代码：002230"; //合成文本

    /* 用户登录 */
    ret = MSPLogin(NULL, NULL, login_params);//第一个参数是用户名，第二个参数是密码，第三个参数是登录参数，用户名和密码可在http://www.xfyun.cn注册获取
    if (MSP_SUCCESS != ret)
    {
        printf("MSPLogin failed, error code: %d.\n", ret);
        goto exit ;//登录失败，退出登录
    }
    printf("\n###########################################################################\n");
    printf("## 语音合成（Text To Speech，TTS）技术能够自动将任意文字实时转换为连续的 ##\n");
    printf("## 自然语音，是一种能够在任何时间、任何地点，向任何人提供语音信息服务的  ##\n");
    printf("## 高效便捷手段，非常符合信息时代海量数据、动态更新和个性化查询的需求。  ##\n");
    printf("###########################################################################\n\n");
    /* 文本合成 */
    printf("开始合成 ...\n");
    ret = text_to_speech(text, filename, session_begin_params);
    if (MSP_SUCCESS != ret)
    {
        printf("text_to_speech failed, error code: %d.\n", ret);
    }
    printf("合成完毕\n");

exit:
    printf("按任意键退出 ...\n");
    getchar();
    MSPLogout(); //退出登录

    return 0;
}

将它修改成ros的节点文件如下:

/*
* 语音合成（Text To Speech，TTS）技术能够自动将任意文字实时转换为连续的
* 自然语音，是一种能够在任何时间、任何地点，向任何人提供语音信息服务的
* 高效便捷手段，非常符合信息时代海量数据、动态更新和个性化查询的需求。
*/

#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <unistd.h>
#include <ros/ros.h>
#include <std_msgs/String.h> 

#include "qtts.h"
#include "msp_cmn.h"
#include "msp_errors.h"

const char* fileName="/home/zc/Music/voice.wav";
const char* playPath="play /home/zc/Music/voice.wav";

/* wav音频头部格式 */
typedef struct _wave_pcm_hdr
{
    char            riff[4];                // = "RIFF"
    int     size_8;                 // = FileSize - 8
    char            wave[4];                // = "WAVE"
    char            fmt[4];                 // = "fmt "
    int     fmt_size;       // = 下一个结构体的大小 : 16

    short int       format_tag;             // = PCM : 1
    short int       channels;               // = 通道数 : 1
    int     samples_per_sec;        // = 采样率 : 8000 | 6000 | 11025 | 16000
    int     avg_bytes_per_sec;      // = 每秒字节数 : samples_per_sec * bits_per_sample / 8
    short int       block_align;            // = 每采样点字节数 : wBitsPerSample / 8
    short int       bits_per_sample;        // = 量化比特数: 8 | 16

    char            data[4];                // = "data";
    int     data_size;              // = 纯数据长度 : FileSize - 44 
} wave_pcm_hdr;

/* 默认wav音频头部数据 */
wave_pcm_hdr default_wav_hdr = 
{
    { 'R', 'I', 'F', 'F' },
    0,
    {'W', 'A', 'V', 'E'},
    {'f', 'm', 't', ' '},
    16,
    1,
    1,
    16000,
    32000,
    2,
    16,
    {'d', 'a', 't', 'a'},
    0  
};
/* 文本合成 */
int text_to_speech(const char* src_text, const char* des_path, const char* params)
{
    int          ret          = -1;
    FILE*        fp           = NULL;
    const char*  sessionID    = NULL;
    unsigned int audio_len    = 0;
    wave_pcm_hdr wav_hdr      = default_wav_hdr;
    int          synth_status = MSP_TTS_FLAG_STILL_HAVE_DATA;

    if (NULL == src_text || NULL == des_path)
    {
        printf("params is error!\n");
        return ret;
    }
    fp = fopen(des_path, "wb");
    if (NULL == fp)
    {
        printf("open %s error.\n", des_path);
        return ret;
    }
    /* 开始合成 */
    sessionID = QTTSSessionBegin(params, &ret);
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSSessionBegin failed, error code: %d.\n", ret);
        fclose(fp);
        return ret;
    }
    ret = QTTSTextPut(sessionID, src_text, (unsigned int)strlen(src_text), NULL);
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSTextPut failed, error code: %d.\n",ret);
        QTTSSessionEnd(sessionID, "TextPutError");
        fclose(fp);
        return ret;
    }
    printf("正在合成 ...\n");
    fwrite(&wav_hdr, sizeof(wav_hdr) ,1, fp); //添加wav音频头，使用采样率为16000
    while (1) 
    {
        /* 获取合成音频 */
        const void* data = QTTSAudioGet(sessionID, &audio_len, &synth_status, &ret);
        if (MSP_SUCCESS != ret)
            break;
        if (NULL != data)
        {
            fwrite(data, audio_len, 1, fp);
            wav_hdr.data_size += audio_len; //计算data_size大小
        }
        if (MSP_TTS_FLAG_DATA_END == synth_status)
            break;
        printf(">");
        usleep(15*1000); //防止频繁占用CPU
    }
    printf("\n");
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSAudioGet failed, error code: %d.\n",ret);
        QTTSSessionEnd(sessionID, "AudioGetError");
        fclose(fp);
        return ret;
    }
    /* 修正wav文件头数据的大小 */
    wav_hdr.size_8 += wav_hdr.data_size + (sizeof(wav_hdr) - 8);

    /* 将修正过的数据写回文件头部,音频文件为wav格式 */
    fseek(fp, 4, 0);
    fwrite(&wav_hdr.size_8,sizeof(wav_hdr.size_8), 1, fp); //写入size_8的值
    fseek(fp, 40, 0); //将文件指针偏移到存储data_size值的位置
    fwrite(&wav_hdr.data_size,sizeof(wav_hdr.data_size), 1, fp); //写入data_size的值
    fclose(fp);
    fp = NULL;
    /* 合成完毕 */
    ret = QTTSSessionEnd(sessionID, "Normal");
    if (MSP_SUCCESS != ret)
    {
        printf("QTTSSessionEnd failed, error code: %d.\n",ret);
    }

    return ret;
}
/* make topic callback to wav file */
void makeTextToWav(const char* text,const char* filename)
{
        int ret=MSP_SUCCESS;
    const char* login_params="appid = 5afcee34, work_dir = .";//登录参数,appid与msc库绑定,请勿随意改动
    /*
    * rdn:           合成音频数字发音方式

    * volume:        合成音频的音量
    * pitch:         合成音频的音调
    * speed:         合成音频对应的语速

    * voice_name:    合成发音人
    * sample_rate:   合成音频采样率
    * text_encoding: 合成文本编码格式

    *
    */
    const char* session_begin_params = "voice_name = xiaowanzi, text_encoding = utf8, sample_rate = 16000, speed = 60, volume = 60, pitch = 50, rdn = 0";
    /* const char* filename             = "tts_sample.wav"; //合成的语音文件名称
    const char* text                 = "大家好，我叫小倩，我今年15岁,我的车牌号是A123456,今天是2018年5月17号"; //合成文本 */

    /* 用户登录 */
    ret = MSPLogin(NULL, NULL, login_params);//第一个参数是用户名，第二个参数是密码，第三个参数是登录参数，用户名和密码可在http://www.xfyun.cn注册获取
    if (MSP_SUCCESS != ret)
    {
        printf("MSPLogin failed, error code: %d.\n", ret);

    }
    /* 文本合成 */
else 
{
    printf("开始合成 ...\n");
    ret = text_to_speech(text, filename, session_begin_params);
    if (MSP_SUCCESS != ret)
    {
        printf("text_to_speech failed, error code: %d.\n", ret);
    }
    printf("合成完毕\n");
}
       MSPLogout();
}
/* play compose.wav file   */
void playWav()
{
system(playPath);
}
/* topic auto invoke,make text to wav file,then play wav file */
void topicCallBack(const std_msgs::String::ConstPtr& msg)
{
std::cout<<"get topic text:"<<msg->data.c_str();

makeTextToWav(msg->data.c_str(),fileName);

playWav();

}
int main(int argc, char* argv[])
{
const char* start = "科大讯飞在线语音合成模块启动成功";
makeTextToWav(start,fileName);
playWav();

    ros::init(argc,argv,"xf_tts_node");
        ros::NodeHandle nd;
        ros::Subscriber sub = nd.subscribe("/voice/xf_tts_topic",3,topicCallBack);
        ros::spin();
    return 0;
}

总的修改思路如下:将main函数里的内容剪切,然后放进另一个自己定义的函数里,那么main函数里面空了.参考发布节点的main函数里的格式,在main函数里面加入初始化,句柄,回调函数,订阅者等内容,另外开头加上ROS的头文件以及消息类型的头文件,具体参考如下wiki上的编写订阅器节点文件:

#include "ros/ros.h"
#include "std_msgs/String.h"

/**
 * This tutorial demonstrates simple receipt of messages over the ROS system.
 */
void chatterCallback(const std_msgs::String::ConstPtr& msg)
{
  ROS_INFO("I heard: [%s]", msg->data.c_str());
}

int main(int argc, char **argv)
{
  /**
   * The ros::init() function needs to see argc and argv so that it can perform
   * any ROS arguments and name remapping that were provided at the command line. For programmatic
   * remappings you can use a different version of init() which takes remappings
   * directly, but for most command-line programs, passing argc and argv is the easiest
   * way to do it.  The third argument to init() is the name of the node.
   *
   * You must call one of the versions of ros::init() before using any other
   * part of the ROS system.
   */
  ros::init(argc, argv, "listener");

  /**
   * NodeHandle is the main access point to communications with the ROS system.
   * The first NodeHandle constructed will fully initialize this node, and the last
   * NodeHandle destructed will close down the node.
   */
  ros::NodeHandle n;

  /**
   * The subscribe() call is how you tell ROS that you want to receive messages
   * on a given topic.  This invokes a call to the ROS
   * master node, which keeps a registry of who is publishing and who
   * is subscribing.  Messages are passed to a callback function, here
   * called chatterCallback.  subscribe() returns a Subscriber object that you
   * must hold on to until you want to unsubscribe.  When all copies of the Subscriber
   * object go out of scope, this callback will automatically be unsubscribed from
   * this topic.
   *
   * The second parameter to the subscribe() function is the size of the message
   * queue.  If messages are arriving faster than they are being processed, this
   * is the number of messages that will be buffered up before beginning to throw
   * away the oldest ones.
   */
  ros::Subscriber sub = n.subscribe("chatter", 1000, chatterCallback);

  /**
   * ros::spin() will enter a loop, pumping callbacks.  With this version, all
   * callbacks will be called from within this thread (the main one).  ros::spin()
   * will exit when Ctrl-C is pressed, or the node is shutdown by the master.
   */
  ros::spin();

  return 0;
}

最后cmakelist里加上要执行的文件以及依赖库

add_executable(xf_tts_node src/xf_tts.cpp)
target_link_libraries(xf_tts_node ${catkin_LIBRARIES} -lmsc -lrt -ldl -lpthread)

以及前面include_directories里加上include

include_directories(
    include

  ${catkin_INCLUDE_DIRS}
)

这样就行了

三.测试运行

开多个终端
第一个终端

roscore

第二个终端

rosrun voice_system xf_tts_node

可以听见科大讯飞在线语音模块启动成功的声音
这时可以打开topic看看有哪些节点在运行,以及手动用命令发布消息让其播报
命令如下:

rostopic list
rostopic info /voice/xf_tts_topic
rostopic pub /voice/xf_tts_topic std_msgs/String "你好"

贴图如下:


这时能听见你好的声音,说明成功了.程序改的没问题.