msb-public
/
PaddleSpeech
mirror of https://github.com/PaddlePaddle/PaddleSpeech
1
0
Code Issues Projects Releases Wiki Activity

fix image link (#612)

Browse Source
pull/619/head
Hui Zhang 3 years ago committed by GitHub
parent db022fac6e
commit 2bdf4c946a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File

2
doc/src/benchmark.md
Unescape View File

@ -4,7 +4,7 @@
We compare the training time with 1, 2, 4, 8 Tesla V100 GPUs (with a subset of LibriSpeech samples whose audio durations are between 6.0 and 7.0 seconds). And it shows that a **near-linear** acceleration with multiple GPUs has been achieved. In the following figure, the time (in seconds) cost for training is printed on the blue bars. We compare the training time with 1, 2, 4, 8 Tesla V100 GPUs (with a subset of LibriSpeech samples whose audio durations are between 6.0 and 7.0 seconds). And it shows that a **near-linear** acceleration with multiple GPUs has been achieved. In the following figure, the time (in seconds) cost for training is printed on the blue bars.
<img src="images/multi_gpu_speedup.png" width=450><br/> <img src="../images/multi_gpu_speedup.png" width=450><br/>
| # of GPU | Acceleration Rate | | # of GPU | Acceleration Rate |
| -------- | --------------: | | -------- | --------------: |

2
doc/src/text_front_end.md
Unescape View File

@ -101,7 +101,7 @@ LP -> LO -> L1(#1) -> L2(#2) -> L3(#3) -> L4(#4) -> L5 -> L6 -> L7
常用方法使用的是级联CRF，首先预测如果是PW，再继续预测是否是PPH，再预测是否是IPH 常用方法使用的是级联CRF，首先预测如果是PW，再继续预测是否是PPH，再预测是否是IPH
<img src="images/prosody.jpeg" width=450><br/> <img src="../images/prosody.jpeg" width=450><br/>
论文: 2015 .Ding Et al. - Automatic Prosody Prediction For Chinese Speech Synthesis Using BLSTM-RNN and Embedding Features 论文: 2015 .Ding Et al. - Automatic Prosody Prediction For Chinese Speech Synthesis Using BLSTM-RNN and Embedding Features

Write Preview
Loading…
Cancel
Save