icefall
icefall copied to clipboard
k2-fsa
I was wondering if there is any plan on adding context graph to k2/Icefall (something similar to :https://wenet.org.cn/wenet/context.html ) cc @csukuangfj
Hi, I'm noticing some weird behaviour when altering the right context length with the conv emformer. Basically, increasing the right context length, while keeping chunk size constant, results in higher...
https://github.com/k2-fsa/icefall/issues/626#issuecomment-1293840023 The above error can be easily caught by CI tests. We should add tests for every released model mentioned in `RESULTS.md`. https://github.com/k2-fsa/icefall/tree/master/.github/scripts is a good start.
I have installed k2, and when testing it I got the expected results. However for icefall, although I did not get any errors, when running the yesno example I got...
I tried to create icefall recipe for MGB2-100h and I did not get any error during the preparation stage. However I got an error during the training after batch 600....
This PR adds a recipe of LSTM model for AISHELL dataset. Refer to https://github.com/k2-fsa/icefall/pull/479 and https://github.com/k2-fsa/icefall/pull/564
[log_prepare_stage14_maxDur500_1GPU_01.txt](https://github.com/k2-fsa/icefall/files/9635055/log_prepare_stage14_maxDur500_1GPU_01.txt) [train.py.txt](https://github.com/k2-fsa/icefall/files/9635057/train.py.txt) I modify Gigiaspeech recipe to train my 1200 hrs data (Including short utterances and long utterances with segmentations). I also encountered the invalid input size as at https://github.com/k2-fsa/icefall/issues/320...
From https://github.com/k2-fsa/icefall/pull/522#issuecomment-1207366968, in decoding script, we tried to introduce `utt_id` into recognition results to better compare: https://github.com/k2-fsa/icefall/blob/a66e74b92f2f41855e0422420dfb06ebe7c6889f/egs/librispeech/ASR/pruned_transducer_stateless5/decode.py#L567 But actually `cut_id` is not the original `utt_id` in dataset, especially after `cut_set.trim_to_supervisions()`,...
Guys, I discovered an issue with the speed of the ActivationBalancer. After implementing NVTX profiling by using `import nvtx` and then using `nvtx.annotate` as both decorators and context managers, I...
Unable to decode AMI SDM full-corpur-asr dev set. 1. Prepared AMI test and dev set using this command: ```bash lhotse prepare ami \ --mic sdm \ --partition full-corpus-asr \ \...
