mirror of
https://github.com/k2-fsa/icefall.git
synced 2025-08-08 09:32:20 +00:00
beab229fd7
* add pruned-rnnt2 recipe for alimeeting dataset * update code for merging * change LilcomHdf5Writer to ChunkedLilcomHdf5Writer * change for test.yml * change for test.yml * change for test.yml * change for workflow yml * change for yml * change for yml * change for README.md * change for yml * solve the conflicts * solve the conflicts
20 lines
908 B
Markdown
20 lines
908 B
Markdown
|
|
# Introduction
|
|
|
|
This recipe includes some different ASR models trained with Alimeeting (far).
|
|
|
|
[./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_stateless2` | 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.
|