-
https://github.com/lfortran/lfortran/issues/3761
```fortran
program main
real(8), allocatable :: mean_
real(8) :: mean_val
allocate(mean_, source = mean_val)
end
```
```console
$…
-
**Problem with decode result on SPGISpeech dataset**
Hi, I downloaded the pretrained model from [https://zenodo.org/record/4585546](https://zenodo.org/record/4585546) and inference with differe…
-
Here are some baseline benchmarks.
I used the lf5 branch, commit 995bffd104013fad37aca8c10ee916b8b4044cb2.
```diff
--- a/src/input.f90
+++ b/src/input.f90
@@ -76,11 +76,11 @@ MODULE input_mod…
-
(dh) (base) root@autodl-container-2d0241b899-289114e4:~/Ultralight-Digital-Human-master# python inference.py --asr wenet --dataset /root/metahumanvideo --audio_feat /root/metahumanvideo/aud_wenet.npy …
-
Is there a way to use this in my app without a Chinese phone number?
i was trying to register on the baidu platform but my number does not work.
and is there a way to send a wav file for recogniztio…
-
Hi. Firstly, thank you for open-sourcing the model and the libraries.
I tried testing the streaming ASR Jupyter [notebook](https://github.com/espnet/notebook/blob/master/espnet2_streaming_asr_demo…
-
Please add the Squeezeformer model architecture for ASR as it is very efficient compared to Conformer.
-
There are some of the functions implemented in `comptime_eval.h` which are then implemented in `intrinsic_function_registry.h`.
For example, `kind` is implemented as an intrinsic now, so it doesn't a…
-
In the paper entitled: "Improving Massively Multilingual ASR with Auxiliary CTC Objectives" you proposed a method where early encoder layers where conditioned to identify the language of speech, while…
-
Hello, I have a question about 10-hour ASR fine-tuning in your paper.
Can you give me a procedure about this experiment? (or the link I can refer)
I just want to conduct the my own experiments for…