-
I was working on scala-collection-compat to add an extension method for `knownSize` and I notice a couple of inconsistencies.
This tables summarize a bunch of collection containers, their `knownSiz…
-
Following up a somewhat strange example in which @dshiber inadvertantly built a 18.6 million observation timeseries, I saw that there is a fat hot spot in two places in LinearIndex.fs, the first where…
-
The following snippet from `ivector-bin/agglomerative-cluster.cc` assigns `threshold` to its negative after reading in each recording. So threshold toggles from, e.g., 0.1, -0.1, 0.1, etc. Also, beca…
-
Hi @jinserk ,
Have you ever encountered the following error? My system is Ubuntu 16.04.5 LTS with Python 3.6.8.
```
python train.py deepspeech_ctc --use-cuda
Traceback (most recent call last):…
-
Our application has multiple concurrent threads each with a Deedle frame. Each frame contains objects of the same type with properties that are not primitive types.
When two threads attempt to expand …
-
请问我在用CAPI实现seq2seq预测模型时,只能得到生成序列的id,但是无法得到生成序列的打分值。请问下原因?
附:预测代码
```cpp
int sentence_ids[] = {0, 33015, 10213, 4, 87713, 12908, 630;
paddle_ivector sentence = paddle_ivector_create(
…
-
We had performance drop during online recognition that included silences. Adding silence-weight option fixes the records that include say up to 1000-1500 ms (not sure about the exact duration, it migh…
-
模型输入结构
```
integer_value_sequence(DICT_SIZE),
dense_vector(1024),
```
输入数据demo
```
[[[1530,222],[0.196966, -1.395368, -0.135187, -0.648595, 0.381318,...]],
[[1530,222],[-5.097437999999999…
-
I currently have a case where two Frames with identical content are considered unequal because, although the rowindices contain the same values in the same order, they indices do not contain exactly t…
-
Hi, sorry to bother with such a question.
I used kaldi to train lre07 to get the models to identify spoken languages.
I want to apply the models for online recognizing languages. And I think asr-se…