mirror of
https://github.com/k2-fsa/icefall.git
synced 2025-08-09 18:12:19 +00:00
5c3ee8bfcd
* pruned-transducer-stateless5 recipe for aishell4 * pruned-transducer-stateless5 recipe for aishell4 * do some changes and text normalize * do some changes * add text normalize * combine the training data and decode without webdataset * update codes for merging * Do a change for READMD.md
20 lines
937 B
Markdown
20 lines
937 B
Markdown
|
|
# Introduction
|
|
|
|
This recipe includes some different ASR models trained with Aishell4 (including S, M and L three subsets).
|
|
|
|
[./RESULTS.md](./RESULTS.md) contains the latest results.
|
|
|
|
# Transducers
|
|
|
|
There are various folders containing the name `transducer` in this folder.
|
|
The following table lists the differences among them.
|
|
|
|
| | Encoder | Decoder | Comment |
|
|
|---------------------------------------|---------------------|--------------------|-----------------------------|
|
|
| `pruned_transducer_stateless5` | Conformer(modified) | Embedding + Conv1d | Using k2 pruned RNN-T loss | |
|
|
|
|
The decoder in `transducer_stateless` is modified from the paper
|
|
[Rnn-Transducer with Stateless Prediction Network](https://ieeexplore.ieee.org/document/9054419/).
|
|
We place an additional Conv1d layer right after the input embedding layer.
|