Update results for torchaudio RNN-T. (#322)

This commit is contained in:
Fangjun Kuang 2022-04-20 11:15:10 +08:00 committed by GitHub
parent fce7f3cd9a
commit 3607c516d6
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -7,7 +7,7 @@ This is with a reworked version of the conformer encoder, with many changes.
#### Training on fulll librispeech
using commit `34aad74a2c849542dd5f6359c9e6b527e8782fd6`.
Using commit `34aad74a2c849542dd5f6359c9e6b527e8782fd6`.
See <https://github.com/k2-fsa/icefall/pull/288>
The WERs are:
@ -353,8 +353,12 @@ You can find a pretrained model by visiting
##### 2022-04-19
[transducer_stateless2](./transducer_stateless2)
This version uses torchaudio's RNN-T loss.
Using commit `fce7f3cd9a486405ee008bcbe4999264f27774a3`.
See <https://github.com/k2-fsa/icefall/pull/316>
| | test-clean | test-other | comment |
|-------------------------------------|------------|------------|--------------------------------------------------------------------------------|
| greedy search (max sym per frame 1) | 2.65 | 6.30 | --epoch 59 --avg 10 --max-duration 600 |