-
I am following the recipe in this link
https://github.com/espnet/espnet/blob/master/egs2/TEMPLATE/tts1/README.md
Everything runs smoothly but in ```exp/tts_train_raw_phn_tacotron_g2p_en_no_space…
-
RNN-Transducer is a recently added prototype module and users can opt-in/out when building torchaudio.
The unit test for RNNT is supposed to be skipped when RNNT is not available, but instead it sk…
-
It seems like the commit 7e3b939237e69a396a017e88bf836972f0d63e79 broke part of my config.
The config looks like this:
```
network = {
"output": {
"class": "rec", "from": "data", # time-…
-
I tried on librispeech with 4 gpu, always CUDA OOM, even if I use a very small model and very small batch.
Is there a recipe I could follow?
thx
-
Hello
I am trying to train subword_conformer with rnnt loss in TF without installing warp-transducer as instructed in the readme but getting the following error. Any help would be appreciated.
Tha…
-
This issue is to collect some thoughts on the recurrent loops design, which wraps the `RecLayer` with an explicit subnetwork in RETURNN.
The main goal is to have this very straight-forward and simp…
-
hi,
Have you ever run a example of conformer transducer, i try to run on aishell but have some problem.
first ,i copy the configuration of librispeech -- lstm_rnnt_bpe1000.yaml and it's ok
then i …
-
I was wondering if you could comment why you're dropping 1st target [here](https://github.com/sooftware/KoSpeech/blob/latest/kospeech/trainer/supervised_trainer.py#L433) when training conformer with …
-
Hi! I have cloned the latest version of the repo and I am running into 2 issues when using the example configs for the streaming transducer model:
1) the `test_streaming_transducer.py` script only ru…
-
**Describe the bug**
When I used aishell data to train a transformer-transducer, 48GB of memory was not enough. In the past, the memory usage was 47909MiB/48600MiB(only espnet training), but today’s …