Instructions to use asafaya/hubert-large-arabic-transcribe with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use asafaya/hubert-large-arabic-transcribe with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="asafaya/hubert-large-arabic-transcribe")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("asafaya/hubert-large-arabic-transcribe") model = AutoModel.from_pretrained("asafaya/hubert-large-arabic-transcribe", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Ali Safaya commited on
Commit ·
80ae584
1
Parent(s): d24bf64
December 22 update
Browse files- hyperparams.yaml +2 -2
- model.ckpt +2 -2
- tokenizer.ckpt +2 -2
- wav2vec2.ckpt +1 -1
hyperparams.yaml
CHANGED
|
@@ -13,7 +13,7 @@ freeze_wav2vec: false
|
|
| 13 |
dropout: 0.2
|
| 14 |
|
| 15 |
# Outputs
|
| 16 |
-
output_neurons:
|
| 17 |
tokenizer: !new:sentencepiece.SentencePieceProcessor
|
| 18 |
|
| 19 |
# Decoding parameters
|
|
@@ -52,7 +52,7 @@ wav2vec2: !new:speechbrain.lobes.models.huggingface_wav2vec.HuggingFaceWav2Vec2
|
|
| 52 |
|
| 53 |
ctc_lin: !new:speechbrain.nnet.linear.Linear
|
| 54 |
input_size: 1024
|
| 55 |
-
n_neurons:
|
| 56 |
|
| 57 |
log_softmax: !new:speechbrain.nnet.activations.Softmax
|
| 58 |
apply_log: true
|
|
|
|
| 13 |
dropout: 0.2
|
| 14 |
|
| 15 |
# Outputs
|
| 16 |
+
output_neurons: 80 # BPE size, index(blank/eos/bos) = 0
|
| 17 |
tokenizer: !new:sentencepiece.SentencePieceProcessor
|
| 18 |
|
| 19 |
# Decoding parameters
|
|
|
|
| 52 |
|
| 53 |
ctc_lin: !new:speechbrain.nnet.linear.Linear
|
| 54 |
input_size: 1024
|
| 55 |
+
n_neurons: 80
|
| 56 |
|
| 57 |
log_softmax: !new:speechbrain.nnet.activations.Softmax
|
| 58 |
apply_log: true
|
model.ckpt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59d5375d484602ef75137633303a63e501c91a49141381651b605a2e9d85ce45
|
| 3 |
+
size 12980350
|
tokenizer.ckpt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29f9c921e0a01f8ff81739929a65bf16417ead180b7d82e105000887f8355434
|
| 3 |
+
size 238839
|
wav2vec2.ckpt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1261933253
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e05705d7b9594d80f25f64d0e075d016cb1722adbd52ff7c52877af132e3d62f
|
| 3 |
size 1261933253
|