-
In this issue you can either:
- Add papers that you think are interesting to read and discuss (please stick to the format).
- **Vote**: should be done using ๐ on comments
Example: [Voting Paperโฆ
-
```python
LightXML/apex$ pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
WARNING: Disabling all use of wheels due to the use of โฆ
-
# ์?
๋
ผ๋ฌธ์ ์คํฐ๋ & ๋ฐํํจ์ ์์ด, ๋
ผ๋ฌธ์ ์ด๋ป๊ฒ ์ฝ๊ณ ์ดํดํ ๊ฒ์ธ์ง์ ๋ํ "ํ"์ ํ์คํ๊ฒ ์ ์ํ๊ณ ์งํํ๋ ๊ฒ์ด ์ข์ ๊ฒ.
- ๋
ผ๋ฌธ์ ๋จ์ ๋ฒ์ญ์ด ์๋ ์ดํด๊ฐ ์ฐ๋ฆฌ์ ๋ชฉํ.
- ๋
ผ๋ฌธ์ ์ดํดํ๋ ํต์ผ๋ ํ์ ์ ์ํ ์ ์๋ค๋ฉด, ๊ทธ๊ฒ์ ์คํฐ๋์๋ค ์ฌ์ด์ ์ํต์๋ ํฐ ๋์์ด ๋ ๊ฒ.
- ์ฌ๋ฌ๋ฒ์ ์ํ์ฐฉ์ค๋ฅผ ํตํด ๊ทธ๋ฐ ํ์ ์ ์ํ ์ โฆ
-
Hi, I try to understand lstm transducer completely, I have successfully migrated your model to a single file, and also able to load the pretrained model, this is the model I got:
lstm-transducer-liโฆ
-
I'm looking to re-export [this](https://k2-fsa.github.io/sherpa/onnx/pretrained_models/online-transducer/lstm-transducer-models.html#csukuangfj-sherpa-onnx-lstm-en-2023-02-17-english) English streaminโฆ
-
Notice: In order to resolve issues more efficiently, please raise issue following the template.
๏ผๆณจๆ๏ผไธบไบๆดๅ ้ซๆ็่งฃๅณๆจ้ๅฐ็้ฎ้ข๏ผ่ฏทๆ็
งๆจกๆฟๆ้ฎ๏ผ่กฅๅ
็ป่๏ผ
## ๐ Bug
ไฝฟ็จ1.x็ๆฌfunasr๏ผ่ทaishell่ฎญ็ปไพๅญๆถ๏ผๅจstage 1 compute_audio_cmvโฆ
-
Does the sherpa server support VAD or endpointing?
-
python 3.8.11
CUDA: V10.2.89
CUDNN: V8.2.1.32
pytorch: 1.9.0+cu102
GPU: GTX1650
Driver Version: 457.49
pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./โฆ
-
After getting help from [this](https://github.com/TensorSpeech/TensorFlowASR/issues/168), I successfully trained a stream-ransducer model and converted the model to tflite.
I tested the trained modโฆ
-
In the `fast_beam_search()` method, the lattice is eventually generated at: https://github.com/k2-fsa/icefall/blob/ffe816e2a8314318a4ef6d5eaba34b62b842ba3f/egs/librispeech/ASR/pruned_transducer_statelโฆ