Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PaddlePaddle
DeepSpeech
提交
d4ffa161
D
DeepSpeech
项目概览
PaddlePaddle
/
DeepSpeech
大约 2 年 前同步成功
通知
210
Star
8425
Fork
1598
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
245
列表
看板
标记
里程碑
合并请求
3
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
D
DeepSpeech
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
245
Issue
245
列表
看板
标记
里程碑
合并请求
3
合并请求
3
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
提交
d4ffa161
编写于
4月 24, 2022
作者:
H
Hui Zhang
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
format
上级
147338ee
变更
9
隐藏空白更改
内联
并排
Showing
9 changed file
with
14 addition
and
10 deletion
+14
-10
paddlespeech/__init__.py
paddlespeech/__init__.py
+0
-2
speechx/examples/ds2_ol/decoder/ctc-prefix-beam-search-decoder-ol.cc
...mples/ds2_ol/decoder/ctc-prefix-beam-search-decoder-ol.cc
+3
-1
speechx/examples/ds2_ol/decoder/wfst-decoder-ol.cc
speechx/examples/ds2_ol/decoder/wfst-decoder-ol.cc
+3
-1
speechx/examples/ds2_ol/feat/linear-spectrogram-wo-db-norm-ol.cc
.../examples/ds2_ol/feat/linear-spectrogram-wo-db-norm-ol.cc
+1
-1
speechx/speechx/decoder/param.h
speechx/speechx/decoder/param.h
+3
-1
speechx/speechx/frontend/audio/audio_cache.h
speechx/speechx/frontend/audio/audio_cache.h
+1
-1
speechx/speechx/frontend/audio/mfcc.cc
speechx/speechx/frontend/audio/mfcc.cc
+1
-1
speechx/speechx/nnet/paddle_nnet.cc
speechx/speechx/nnet/paddle_nnet.cc
+1
-1
speechx/speechx/nnet/paddle_nnet.h
speechx/speechx/nnet/paddle_nnet.h
+1
-1
未找到文件。
paddlespeech/__init__.py
浏览文件 @
d4ffa161
...
...
@@ -14,5 +14,3 @@
import
_locale
_locale
.
_getdefaultlocale
=
(
lambda
*
args
:
[
'en_US'
,
'utf8'
])
speechx/examples/ds2_ol/decoder/ctc-prefix-beam-search-decoder-ol.cc
浏览文件 @
d4ffa161
...
...
@@ -41,7 +41,9 @@ DEFINE_string(
DEFINE_string
(
model_output_names
,
"softmax_0.tmp_0,tmp_5,concat_0.tmp_0,concat_1.tmp_0"
,
"model output names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_shapes
,
"5-1-1024,5-1-1024"
,
"model cache shapes"
);
using
kaldi
::
BaseFloat
;
...
...
speechx/examples/ds2_ol/decoder/wfst-decoder-ol.cc
浏览文件 @
d4ffa161
...
...
@@ -44,7 +44,9 @@ DEFINE_string(
DEFINE_string
(
model_output_names
,
"softmax_0.tmp_0,tmp_5,concat_0.tmp_0,concat_1.tmp_0"
,
"model output names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_shapes
,
"5-1-1024,5-1-1024"
,
"model cache shapes"
);
using
kaldi
::
BaseFloat
;
...
...
speechx/examples/ds2_ol/feat/linear-spectrogram-wo-db-norm-ol.cc
浏览文件 @
d4ffa161
...
...
@@ -42,7 +42,7 @@ int main(int argc, char* argv[]) {
int32
num_done
=
0
,
num_err
=
0
;
// feature pipeline: wave cache --> hanning window
// feature pipeline: wave cache --> hanning window
// -->linear_spectrogram --> global cmvn -> feat cache
std
::
unique_ptr
<
ppspeech
::
FrontendInterface
>
data_source
(
...
...
speechx/speechx/decoder/param.h
浏览文件 @
d4ffa161
...
...
@@ -43,7 +43,9 @@ DEFINE_string(
DEFINE_string
(
model_output_names
,
"softmax_0.tmp_0,tmp_5,concat_0.tmp_0,concat_1.tmp_0"
,
"model output names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_names
,
"chunk_state_h_box,chunk_state_c_box"
,
"model cache names"
);
DEFINE_string
(
model_cache_shapes
,
"5-1-1024,5-1-1024"
,
"model cache shapes"
);
...
...
speechx/speechx/frontend/audio/audio_cache.h
浏览文件 @
d4ffa161
...
...
@@ -58,7 +58,7 @@ class AudioCache : public FrontendInterface {
std
::
mutex
mutex_
;
std
::
condition_variable
ready_feed_condition_
;
kaldi
::
int32
timeout_
;
// millisecond
bool
to_float32_
;
// int16 -> float32. used in linear_spectrogram
bool
to_float32_
;
// int16 -> float32. used in linear_spectrogram
DISALLOW_COPY_AND_ASSIGN
(
AudioCache
);
};
...
...
speechx/speechx/frontend/audio/mfcc.cc
浏览文件 @
d4ffa161
...
...
@@ -30,7 +30,7 @@ using kaldi::Matrix;
using
std
::
vector
;
Mfcc
::
Mfcc
(
const
MfccOptions
&
opts
,
std
::
unique_ptr
<
FrontendInterface
>
base_extractor
)
std
::
unique_ptr
<
FrontendInterface
>
base_extractor
)
:
opts_
(
opts
),
computer_
(
opts
.
mfcc_opts
),
window_function_
(
computer_
.
GetFrameOptions
())
{
...
...
speechx/speechx/nnet/paddle_nnet.cc
浏览文件 @
d4ffa161
...
...
@@ -206,7 +206,7 @@ void PaddleNnet::FeedForward(const Vector<BaseFloat>& features,
inferences
->
Resize
(
row
*
col
);
*
inference_dim
=
col
;
output_tensor
->
CopyToCpu
(
inferences
->
Data
());
ReleasePredictor
(
predictor
);
}
...
...
speechx/speechx/nnet/paddle_nnet.h
浏览文件 @
d4ffa161
...
...
@@ -24,7 +24,7 @@ namespace ppspeech {
struct
ModelOptions
{
std
::
string
model_path
;
std
::
string
param_path
;
int
thread_num
;
// predictor thread pool size
int
thread_num
;
// predictor thread pool size
bool
use_gpu
;
bool
switch_ir_optim
;
std
::
string
input_names
;
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录