Automatic Speech Recognition
Transformers
Safetensors
English
asr_model
feature-extraction
asr
speech-recognition
audio
qwen
glm-asr
custom_code
Instructions to use mazesmazes/tiny-audio-next-multiasr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mazesmazes/tiny-audio-next-multiasr with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="mazesmazes/tiny-audio-next-multiasr", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mazesmazes/tiny-audio-next-multiasr", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Training in progress, step 2000
Browse files- config.json +4 -1
- model.safetensors +1 -1
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
"architectures": [
|
| 3 |
"ASRModel"
|
| 4 |
],
|
| 5 |
-
"attn_implementation": "
|
| 6 |
"audio_config": {
|
| 7 |
"_name_or_path": "zai-org/GLM-ASR-Nano-2512",
|
| 8 |
"architectures": [
|
|
@@ -237,6 +237,8 @@
|
|
| 237 |
"eos_token_id": 151645,
|
| 238 |
"freeze_language_model": false,
|
| 239 |
"freeze_projector": false,
|
|
|
|
|
|
|
| 240 |
"length_penalty": 1.0,
|
| 241 |
"llm_dim": 1024,
|
| 242 |
"lora_alpha": 32,
|
|
@@ -262,6 +264,7 @@
|
|
| 262 |
"pad_token_id": 151643,
|
| 263 |
"pipeline_tag": "automatic-speech-recognition",
|
| 264 |
"pretrained_model_path": "mazesmazes/tiny-audio-next-multiasr",
|
|
|
|
| 265 |
"projector_hidden_dim": 1024,
|
| 266 |
"projector_pool_stride": 4,
|
| 267 |
"projector_type": "mlp",
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"ASRModel"
|
| 4 |
],
|
| 5 |
+
"attn_implementation": "flash_attention_2",
|
| 6 |
"audio_config": {
|
| 7 |
"_name_or_path": "zai-org/GLM-ASR-Nano-2512",
|
| 8 |
"architectures": [
|
|
|
|
| 237 |
"eos_token_id": 151645,
|
| 238 |
"freeze_language_model": false,
|
| 239 |
"freeze_projector": false,
|
| 240 |
+
"freeze_text_embed_tokens": false,
|
| 241 |
+
"label_smoothing": 0.1,
|
| 242 |
"length_penalty": 1.0,
|
| 243 |
"llm_dim": 1024,
|
| 244 |
"lora_alpha": 32,
|
|
|
|
| 264 |
"pad_token_id": 151643,
|
| 265 |
"pipeline_tag": "automatic-speech-recognition",
|
| 266 |
"pretrained_model_path": "mazesmazes/tiny-audio-next-multiasr",
|
| 267 |
+
"projector_dropout": 0.1,
|
| 268 |
"projector_hidden_dim": 1024,
|
| 269 |
"projector_pool_stride": 4,
|
| 270 |
"projector_type": "mlp",
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2408324496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a68edaf462dc0dff106ee98009731f2fe32fa8e7ff830db243c3843f3f04d55
|
| 3 |
size 2408324496
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5393
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93249932015f989823ad2ba998cfe00b3cf2b93f3e7532c45177ee0fb8635ad2
|
| 3 |
size 5393
|