-
## 简介
non-autogression decoding. 用iterative的方式不断refine已有的翻译结果。第一轮是完全NAT，然后再次基础上，选择confidence差的N个词，mask掉，去对mask的内容进行refine，迭代X轮结束。对mask的恢复过程类似BERT。整篇文章读起来很舒服，听起来也合理。后面实验部分也很饱满。
## 论文信息
* Author: F…
-
**Description:**
Predicting future traffic flow, which will aid in traffic management and planning. The goal is to build a model that can accurately forecast traffic flow based on historical data a…
-
Hello,
I trained an ASR end-to-end system based on MaskCTC. The quality of speech recognition is great and it is amazing how fast the model can decode speech. However, I was wondering if there is a…
-
## Why
Machine Learning 輪講は最新の技術や論文を追うことで、エンジニアが「技術で解決できること」のレベルをあげていくことを目的にした会です。
prev. #19
## What
話したいことがある人はここにコメントしましょう！
面白いものを見つけた時点でとりあえず話すという宣言だけでもしましょう！
-
This issue is to track how to get German working and what options one need to consider.
# dotnet-examples
https://github.com/k2-fsa/sherpa-onnx/tree/master/dotnet-examples
- [ ] keyword-spotting-…
-
Currently, in the TimeGrad model, experimental results reveal suboptimal adherence to control inputs when using covariate conditioning. This issue stems from the dominance of previous sequences within…
-
They claim lookahead decoding provides a 1.5~2x decoding speedup without a speculative model.
Blog post: https://lmsys.org/blog/2023-11-21-lookahead-decoding/
Twitter thread: https://twitter.com/l…
-
In the paper, only two contrasting decoding methods were shown: AR (one character at a time) and non-AR (all characters at once). But in theory, PARSeq supports semi-autoregressive decoding and anythi…
baudm updated
2 years ago
-
### 🚀 The feature, motivation and pitch
I calculated in non-autoregressive tts, asr, machine-translation using transformer with convolutional network. Machine learning was well carried out. So, I …
-
The code initially indicates that CUDA will be disabled but continues to attempt its use later, resulting in failures. The intention is to run the code exclusively on the CPU.
Code:
def init…