parent
9c09837915
commit
d983f8d8b3
@ -0,0 +1,17 @@
|
|||||||
|
# Model Arcitecture
|
||||||
|
|
||||||
|
The implemented arcitecure of Deepspeech2 online model is based on [Deepspeech2 model](https://arxiv.org/pdf/1512.02595.pdf) with some changes.
|
||||||
|
The figure of arcitecture is shown in ![image](../image/ds2onlineModel.png).
|
||||||
|
The model is mainly composed of 2D convolution subsampling layer and single direction rnn layers. To illustrate the model implementation in detail, 5 parts is introduced.
|
||||||
|
1. Feature Extraction.
|
||||||
|
2. 2D Convolution subsampling layer.
|
||||||
|
3. RNN layer with only forward direction.
|
||||||
|
4. Softmax Layer.
|
||||||
|
5. CTC Decoder.
|
||||||
|
|
||||||
|
|
||||||
|
# Feature Extraction
|
||||||
|
|
||||||
|
Three methods of feature extraction is implemented, which are linear, fbank and mfcc.
|
||||||
|
For a single utterance $x^i$ sampled from the training set $S$,
|
||||||
|
$ S= {(x^1,y^1),(x^2,y^2),...,(x^m,y^m)}$, where $y^i$ is the label correspodding to the ${x^i}
|
Loading…
Reference in new issue