smajumdar commited on
Commit
acda77c
1 Parent(s): d1999eb

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +93 -0
README.md CHANGED
@@ -1,3 +1,96 @@
1
  ---
 
 
2
  license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - kr
4
  license: apache-2.0
5
+ library_name: nemo
6
+ datasets:
7
+ - ksponspeech
8
+ thumbnail: null
9
+ tags:
10
+ - automatic-speech-recognition
11
+ - speech
12
+ - audio
13
+ - RNNT
14
+ - Conformer
15
+ - Transformer
16
+ - NeMo
17
+ - pytorch
18
+ model-index:
19
+ - name: stt_kr_conformer_transducer_large
20
+ results: []
21
+
22
  ---
23
+
24
+
25
+ ## Model Overview
26
+
27
+ <DESCRIBE IN ONE LINE THE MODEL AND ITS USE>
28
+
29
+ ## NVIDIA NeMo: Training
30
+
31
+ To train, fine-tune or play with the model you will need to install [NVIDIA NeMo](https://github.com/NVIDIA/NeMo). We recommend you install it after you've installed latest Pytorch version.
32
+ ```
33
+ pip install nemo_toolkit['all']
34
+ ```
35
+
36
+ ## How to Use this Model
37
+
38
+ The model is available for use in the NeMo toolkit [3], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.
39
+
40
+ ### Automatically instantiate the model
41
+
42
+ ```python
43
+ import nemo.collections.asr as nemo_asr
44
+ asr_model = nemo_asr.models.ASRModel.from_pretrained("eesungkim/stt_kr_conformer_transducer_large")
45
+ ```
46
+
47
+ ### Transcribing using Python
48
+ ```
49
+ asr_model.transcribe(['<File name>'])
50
+ ```
51
+
52
+ ### Transcribing many audio files
53
+
54
+ ```shell
55
+ python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py pretrained_name="eesungkim/stt_kr_conformer_transducer_large" audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"
56
+ ```
57
+
58
+ ### Input
59
+
60
+ This model accepts 16000 KHz Mono-channel Audio (wav files) as input.
61
+
62
+ ### Output
63
+
64
+ This model provides transcribed speech as a string for a given audio sample.
65
+
66
+ ## Model Architecture
67
+
68
+ <ADD SOME INFORMATION ABOUT THE ARCHITECTURE>
69
+
70
+ ## Training
71
+
72
+ <ADD INFORMATION ABOUT HOW THE MODEL WAS TRAINED - HOW MANY EPOCHS, AMOUNT OF COMPUTE ETC>
73
+
74
+ ### Datasets
75
+
76
+ <LIST THE NAME AND SPLITS OF DATASETS USED TO TRAIN THIS MODEL (ALONG WITH LANGUAGE AND ANY ADDITIONAL INFORMATION)>
77
+
78
+ ## Performance
79
+
80
+ <LIST THE SCORES OF THE MODEL -
81
+ OR
82
+ USE THE Hugging Face Evaluate LiBRARY TO UPLOAD METRICS>
83
+
84
+ ## Limitations
85
+
86
+ <DECLARE ANY POTENTIAL LIMITATIONS OF THE MODEL>
87
+
88
+ Eg:
89
+ Since this model was trained on publically available speech datasets, the performance of this model might degrade for speech which includes technical terms, or vernacular that the model has not been trained on. The model might also perform worse for accented speech.
90
+
91
+
92
+ ## References
93
+
94
+ <ADD ANY REFERENCES HERE AS NEEDED>
95
+
96
+ [1] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)