RuslanSel
RuslanSel
I ran custom dataset with librispeech/s0/run.sh and conf/train_conformer.yaml and CV loss oscillates near 140 from start without improvement. What could be the reason for this? Thanks in advance.
Low loss
I ran the new version of gigaspeech/run.sh and the loss quickly dropped from 2021-11-17 09:27:17,775 DEBUG TRAIN Batch 0/0 loss 68.487236 loss_att 41.765869 loss_ctc 130.837097 lr 0.00000001 rank 5 2021-11-17...
When I'm run python examples/mms/asr/infer/mms_infer.py --model "/path/to/asr/model" --lang lang_code --audio "/path/to/audio_1.wav" "/path/to/audio_1.wav" I got this error: >>> preparing tmp manifest dir ... >>> loading model & running inference ... File...
In parse_hyp_and_timestamp() after tokens = sp.id_to_piece(res.hyps[i]) I have such tokens: ▁', '', '-', 'mo', 'du', 'le', then words = sp.decode_pieces(tokens).split() gives me two words instead of one '⁇', '-module' but...
Ubuntu 18.04 python 3.11.5 conda pip install --upgrade git+https://github.com/NVIDIA/TransformerEngine.git@stable ... ... [ 79%] Building CXX object common/CMakeFiles/transformer_engine.dir/util/cuda_driver.cpp.o /tmp/pip-req-build-j223vg5e/transformer_engine/common/util/cuda_driver.cpp:8:10: fatal error: filesystem: No such file or directory #include ^~~~~~~~~~~~ Thanks in...