Downloads · 30 days
0
icefall-asr-alimeeting-pruned-transducer-stateless7
desh2608/icefall-asr-alimeeting-pruned-transducer-stateless7
icefall-asr-alimeeting-pruned-transducer-stateless7 is a machine learning model from desh2608. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for k2. The card lists the license as apache-2.0.
This is an ASR recipe for the AliMeeting corpus. AliMeeting provides recordings from the speaker's headset microphones and an 8-channel microphone array. We pool data in the following 4 ways and train a single model o…
Access
Public
Updated Dec 12, 2022
Repo size
611 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt611 MB · 88%
At a glance
- Library
- k2
- License
- apache-2.0
- Access
- Public
- Created
- Dec 9, 2022
- Updated
- Dec 12, 2022
- SHA
- d2739a0b
Datasets
Model card
From the Hugging Face model README
AliMeeting
This is an ASR recipe for the AliMeeting corpus. AliMeeting provides recordings from the speaker's headset microphones and an 8-channel microphone array. We pool data in the following 4 ways and train a single model on the pooled data:
(i) individual headset microphone (IHM) (ii) IHM with simulated reverb (iii) Single distant microphone (SDM) (iv) GSS-enhanced array microphones
Speed perturbation and MUSAN noise augmentation are additionally performed on the pooled data.
Performance Record
pruned_transducer_stateless7
The following are decoded using modified_beam_search:
| Evaluation set | eval CER | test CER |
|---|---|---|
| IHM | 9.58 | 11.53 |
| SDM | 23.37 | 25.85 |
| MDM (GSS-enhanced) | 11.82 | 14.22 |
See the recipe for details.
- Library
- k2
- License
- apache-2.0
- Languages
- zh
- Created
- Dec 9, 2022
- Updated
- Dec 12, 2022