From 4f8bce944e273e317cb84c7046ea514b9d958b4b Mon Sep 17 00:00:00 2001
From: zhuzizyf <42790740+zhuzizyf@users.noreply.github.com>
Date: 星期六, 22 四月 2023 14:54:49 +0800
Subject: [PATCH] Update FsmnVad.cc
---
funasr/runtime/onnxruntime/include/Audio.h | 20 +++++++++++++++-----
1 files changed, 15 insertions(+), 5 deletions(-)
diff --git a/funasr/runtime/onnxruntime/include/Audio.h b/funasr/runtime/onnxruntime/include/Audio.h
index af8d2a9..2667c31 100644
--- a/funasr/runtime/onnxruntime/include/Audio.h
+++ b/funasr/runtime/onnxruntime/include/Audio.h
@@ -2,9 +2,13 @@
#ifndef AUDIO_H
#define AUDIO_H
-#include <ComDefine.h>
#include <queue>
#include <stdint.h>
+#include "Model.h"
+
+#ifndef WAV_HEADER_SIZE
+#define WAV_HEADER_SIZE 44
+#endif
using namespace std;
@@ -20,7 +24,7 @@
~AudioFrame();
int set_start(int val);
- int set_end(int val, int max_len);
+ int set_end(int val);
int get_start();
int get_len();
int disp();
@@ -32,7 +36,6 @@
int16_t *speech_buff;
int speech_len;
int speech_align_len;
- int16_t sample_rate;
int offset;
float align_size;
int data_type;
@@ -43,11 +46,18 @@
Audio(int data_type, int size);
~Audio();
void disp();
- bool loadwav(const char *filename);
+ bool loadwav(const char* filename, int32_t* sampling_rate);
+ void wavResample(int32_t sampling_rate, const float *waveform, int32_t n);
+ bool loadwav(const char* buf, int nLen, int32_t* sampling_rate);
+ bool loadpcmwav(const char* buf, int nFileLen, int32_t* sampling_rate);
+ bool loadpcmwav(const char* filename, int32_t* sampling_rate);
int fetch_chunck(float *&dout, int len);
int fetch(float *&dout, int &len, int &flag);
void padding();
- void split();
+ void split(Model* pRecogObj);
+ float get_time_len();
+
+ int get_queue_size() { return (int)frame_queue.size(); }
};
#endif
--
Gitblit v1.9.1