From 73e410afb9eda1dfb4ff8a6f86ad405ab3e8c48d Mon Sep 17 00:00:00 2001
From: lyblsgo <lyblsgo@163.com>
Date: 星期五, 21 四月 2023 11:00:45 +0800
Subject: [PATCH] rm fftw deps in readme
---
funasr/runtime/onnxruntime/include/Audio.h | 22 ++++++++++++++++++----
1 files changed, 18 insertions(+), 4 deletions(-)
diff --git a/funasr/runtime/onnxruntime/include/Audio.h b/funasr/runtime/onnxruntime/include/Audio.h
index af8d2a9..c38c31a 100644
--- a/funasr/runtime/onnxruntime/include/Audio.h
+++ b/funasr/runtime/onnxruntime/include/Audio.h
@@ -5,6 +5,14 @@
#include <ComDefine.h>
#include <queue>
#include <stdint.h>
+#include "Model.h"
+
+#ifndef model_sample_rate
+#define model_sample_rate 16000
+#endif
+#ifndef WAV_HEADER_SIZE
+#define WAV_HEADER_SIZE 44
+#endif
using namespace std;
@@ -20,7 +28,7 @@
~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 +40,6 @@
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 +50,18 @@
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