-
Proposal to add format preserving cipher modes and supporting these for at least AES. The relevant standard is NIST [SP 800-38G](http://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-38G.p…
-
-
我尝试自己生成了数据集,运行analyzer.py生成了在训练时reader.py中reader所需的file_name_word。
在运行时发生了如下错误,
File "model.py", line 86, in build_attention_state
self.attention, self.time, self.alignments, tuple([]))
Type…
-
-
**Reported by Shaimaa on 2014-03-24 14:20**
When using Microsoft Powerpoint 2010, NVDA doesn't report the alignment information of the text by pressing NVDA+F even when the alignment check box is chec…
-
### Steps to reproduce:
create a text in word, format this text with some information
for example:
bold, italic, centered, font size larger than the word default, font color different from the wo…
-
**What is the background of this bug report?**
When linking a word to a word that is near the end of an utterance, the texts vertically align and the text that proceed the linked word get cut off.
…
-
Hi,
I was wondering whether this is the final code base of your project because I ran into several errors trying to run `run_align.sh`. They stem from https://github.com/sufenlp/AccAlign/blob/84f9b…
-
`Text` should provide the actual metrics of its text. These new out-properties could be named `text-x`, `text-y`, `text-width`, and `text-height`.
With these, you can, e.g., implement a…
-
See notebook, section "Word-level timestamps using attention weights":
https://github.com/openai/whisper/blob/main/notebooks/Multilingual_ASR.ipynb