Downloads · 30 days
13
7% of all-time downloads
visualears-xlsr-300m-v2-cont-step1568
Reza2kn/visualears-xlsr-300m-v2-cont-step1568
visualears-xlsr-300m-v2-cont-step1568 is a automatic speech recognition model from Reza2kn. Use it when you need speech turned into text. It is set up for transformers.
Final continuation checkpoint from /workspace/visualearstrain/runs/visualearsv2contfromckpt1000singleb24ga420260608T050121Z.
All-time downloads
190
Public
Parameters
316M
2.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.8 GB · 99%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- Model type
- wav2vec2
- Access
- Public
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026
- SHA
- b41f6dce
Model card
From the Hugging Face model README
VisualEars XLS-R 300M v2 continuation step 1568
Final continuation checkpoint from /workspace/visualears_train/runs/visualears_v2_cont_from_ckpt1000_single_b24ga4_20260608T050121Z.
Training lineage:
- Continued from DDP checkpoint-1000 of
visualears_v2_decode_gated_alltiers_equal_headkeep_tokfix_b12ga4_ddpsafe_20260608T021656Z. - Single-GPU continuation: batch 24, grad accumulation 4, LR 4e-6, max_steps 568.
Double eval already run before upload:
- Gold-69 WER: 27.69%, CER: 9.62%
- FLEURS fa_ir test WER: 19.15%, CER: 5.84%
model.safetensors SHA256: 085fca58e419e4860d0435b0e31f4d7f484ff2f033915406cf7bf930791e3028
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- wav2vec2
- Languages
- fa
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026