Feature Extraction
Transformers
Safetensors
English
custom_model
multi-modal
conversational
speechllm
speech2text
custom_code
shangeth commited on
Commit
54ac8cb
1 Parent(s): 1098244

Upload model

Browse files
Files changed (3) hide show
  1. README.md +7 -7
  2. config.json +1 -6
  3. config.py +9 -0
README.md CHANGED
@@ -1,19 +1,19 @@
1
  ---
2
- library_name: transformers
3
  language:
4
  - en
5
- metrics:
6
- - wer
7
  license: apache-2.0
8
- datasets:
9
- - librispeech_asr
10
- - mozilla-foundation/common_voice_16_1
11
- - DynamicSuperb/EmotionalSpeechAudioClassification_RAVDESS-EmotionalSound
12
  tags:
13
  - multi-modal
14
  - conversational
15
  - speechllm
16
  - speech2text
 
 
 
 
 
 
17
  ---
18
 
19
  # Model Card for Model ID
 
1
  ---
 
2
  language:
3
  - en
 
 
4
  license: apache-2.0
5
+ library_name: transformers
 
 
 
6
  tags:
7
  - multi-modal
8
  - conversational
9
  - speechllm
10
  - speech2text
11
+ datasets:
12
+ - librispeech_asr
13
+ - mozilla-foundation/common_voice_16_1
14
+ - DynamicSuperb/EmotionalSpeechAudioClassification_RAVDESS-EmotionalSound
15
+ metrics:
16
+ - wer
17
  ---
18
 
19
  # Model Card for Model ID
config.json CHANGED
@@ -1,14 +1,9 @@
1
  {
2
- "architectures": [
3
- "CustomModel"
4
- ],
5
  "audio_enc_dim": 1280,
6
  "auto_map": {
7
- "AutoConfig": "MyConfig.CustomModelConfig",
8
- "AutoModel": "MyModel.CustomModel"
9
  },
10
  "llm_dim": 2048,
11
  "model_type": "custom_model",
12
- "torch_dtype": "float32",
13
  "transformers_version": "4.38.2"
14
  }
 
1
  {
 
 
 
2
  "audio_enc_dim": 1280,
3
  "auto_map": {
4
+ "AutoConfig": "config.SpeechLLMModelConfig"
 
5
  },
6
  "llm_dim": 2048,
7
  "model_type": "custom_model",
 
8
  "transformers_version": "4.38.2"
9
  }
config.py ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ from transformers import PretrainedConfig
2
+
3
+ class SpeechLLMModelConfig(PretrainedConfig):
4
+ model_type = "custom_model"
5
+
6
+ def __init__(self, audio_enc_dim=1280, llm_dim=2048, **kwargs):
7
+ super().__init__(**kwargs)
8
+ self.audio_enc_dim = audio_enc_dim
9
+ self.llm_dim = llm_dim