lijialudew commited on
Commit
aa50616
1 Parent(s): a548b2a

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +54 -0
README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: openrail
3
+ language:
4
+ - en
5
+ metrics:
6
+ - f1
7
+ - recall
8
+ - accuracy
9
+ library_name: speechbrain
10
+ pipeline_tag: audio-classification
11
+ ---
12
+ # Model Card for Model ID
13
+
14
+ <!-- Provide a quick summary of what the model is/does. -->
15
+
16
+ We build a CTC-based ASR model using wav2vec 2.0 (W2V2) for children under 4-year-old. We use two-level fine-tuning to gradually reduce age mismatch between adult ASR to child ASR.
17
+ We first fine-tune W2V2-LibriSpeech960h using [My Science Tutor](https://boulderlearning.com/products/myst/) corpus (consists of conversational speech of students between the third and fifth grades with a virtual tutor) on character level. Then we fine-tune W2V2-MyST using [Providence](https://phonbank.talkbank.org/access/Eng-NA/Providence.html) corpus (consists of longititude audio of 6 English-speaking children aged from 1-4 years interacting with their mothers at home) on phoneme sequences or consonant/vowel sequences.
18
+ We show W2V2-Providence is helpful for improving children's vocalization classification task on two corpus, including [Rapid-ABC](https://openaccess.thecvf.com/content_cvpr_2013/html/Rehg_Decoding_Childrens_Social_2013_CVPR_paper.html) and [BabbleCor](https://osf.io/rz4tx/).
19
+
20
+ ## Model Sources
21
+ For more information regarding this model, please checkout our paper
22
+ - **Paper:** Coming soon
23
+
24
+ ## Model Description
25
+
26
+ <!-- Provide a longer summary of what this model is. -->
27
+ Folder contains the best checkpoint of the following setting
28
+ - **W2V2-MyST by fine-tuning on Librispeech 960h**: save_960h/wav2vec2.ckpt
29
+ - **W2V2-Pro trained on phone sequence**: save_MyST_Providence_ep45_filtered/wav2vec2.ckpt
30
+ - **W2V2-Pro trained on consonant/vowel sequence**: save_MyST_Providence_ep45_filtered_cv_only/wav2vec2.ckpt
31
+
32
+ ## Uses
33
+ **We develop our complete fine-tuning recipe using SpeechBrain toolkit available at**
34
+
35
+ - **https://github.com/jialuli3/speechbrain/tree/infant-voc-classification/recipes/RABC** (used for Rapid-ABC corpus)
36
+ - **https://github.com/jialuli3/speechbrain/tree/infant-voc-classification/recipes/BabbleCor** (used for BabbleCor corpus)
37
+
38
+ # Paper/BibTex Citation
39
+
40
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
41
+ If you found this model helpful to you, please cite us as
42
+
43
+ Coming soon
44
+ <!-- <pre><code>
45
+ </code></pre> -->
46
+
47
+ # Model Card Contact
48
+ Jialu Li (she, her, hers)
49
+
50
+ Ph.D candidate @ Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign
51
+
52
+ E-mail: jialuli3@illinois.edu
53
+
54
+ Homepage: https://sites.google.com/view/jialuli/