Commit
•
b717995
1
Parent(s):
36c6e03
Update README
Browse filesAdd data description.
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- librispeech_asr
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
metrics:
|
| 8 |
+
- wer
|
| 9 |
+
pipeline_tag: automatic-speech-recognition
|
| 10 |
+
tags:
|
| 11 |
+
- asr
|
| 12 |
+
- rescoring
|
| 13 |
+
- rnn-t
|
| 14 |
+
- gpt2
|
| 15 |
+
- nemo
|
| 16 |
+
- lstm
|
| 17 |
+
- kenlm
|
| 18 |
---
|
| 19 |
+
|
| 20 |
+
The data is used in project https://github.com/Alexander92-cpu/LanguageModel_Fusion
|
| 21 |
+
|
| 22 |
+
Data desciption:
|
| 23 |
+
'asr/stt_en_conformer_transducer_small.nemo' - NeMo ASR pre-trained RNN-T model;
|
| 24 |
+
'gpt2' - fine-tuned GPT-2 LM model for rescoring;
|
| 25 |
+
'kenlm/4_ngram_output.bin' - 4-gram language model;
|
| 26 |
+
'lstm' - trained from scratch word-level LSTM LM model and the corresponding tokenizer;
|
| 27 |
+
'text' - contains text data used for training, validation, and testing.
|