Automatic Speech Recognition
Transformers
Safetensors
fun_asr_nano
text-generation
speech-recognition
asr
end-to-end
multilingual
streaming
Instructions to use FunAudioLLM/Fun-ASR-Nano-2512-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FunAudioLLM/Fun-ASR-Nano-2512-hf with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="FunAudioLLM/Fun-ASR-Nano-2512-hf")# Load model directly from transformers import AutoModelForSeq2SeqLM model = AutoModelForSeq2SeqLM.from_pretrained("FunAudioLLM/Fun-ASR-Nano-2512-hf", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Refresh checkpoint for shared Whisper component layout
#2
by langgz - opened
- config.json +14 -14
- generation_config.json +1 -1
- model.safetensors +2 -2
- processor_config.json +0 -2
- tokenizer_config.json +1 -1
config.json
CHANGED
|
@@ -1,28 +1,28 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
"adaptor_dropout": 0.0,
|
| 4 |
"adaptor_intermediate_size": 2048,
|
| 5 |
"adaptor_num_attention_heads": 8,
|
| 6 |
"adaptor_num_hidden_layers": 2,
|
| 7 |
"architectures": [
|
| 8 |
"FunAsrNanoForConditionalGeneration"
|
| 9 |
],
|
| 10 |
-
"
|
| 11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
"dropout": 0.1,
|
|
|
|
|
|
|
| 13 |
"encoder_layers": 50,
|
| 14 |
-
"initializer_range": 0.02,
|
| 15 |
-
"input_size": 560,
|
| 16 |
-
"intermediate_size": 2048,
|
| 17 |
"kernel_size": 11,
|
| 18 |
"model_type": "fun_asr_nano_encoder",
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"tp_blocks": 20
|
| 23 |
},
|
| 24 |
-
"audio_token_index": 151646,
|
| 25 |
-
"dtype": "bfloat16",
|
| 26 |
"initializer_range": 0.02,
|
| 27 |
"model_type": "fun_asr_nano",
|
| 28 |
"text_config": {
|
|
@@ -88,5 +88,5 @@
|
|
| 88 |
"vocab_size": 151936
|
| 89 |
},
|
| 90 |
"tie_word_embeddings": true,
|
| 91 |
-
"transformers_version": "5.
|
| 92 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"activation_function": "relu",
|
|
|
|
| 3 |
"adaptor_intermediate_size": 2048,
|
| 4 |
"adaptor_num_attention_heads": 8,
|
| 5 |
"adaptor_num_hidden_layers": 2,
|
| 6 |
"architectures": [
|
| 7 |
"FunAsrNanoForConditionalGeneration"
|
| 8 |
],
|
| 9 |
+
"audio_token_id": 151646,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"encoder_config": {
|
| 12 |
+
"activation_dropout": 0.1,
|
| 13 |
+
"activation_function": "relu",
|
| 14 |
+
"attention_dropout": 0.1,
|
| 15 |
+
"d_model": 512,
|
| 16 |
"dropout": 0.1,
|
| 17 |
+
"encoder_attention_heads": 4,
|
| 18 |
+
"encoder_ffn_dim": 2048,
|
| 19 |
"encoder_layers": 50,
|
|
|
|
|
|
|
|
|
|
| 20 |
"kernel_size": 11,
|
| 21 |
"model_type": "fun_asr_nano_encoder",
|
| 22 |
+
"num_mel_bins": 80,
|
| 23 |
+
"num_stacked_frames": 7,
|
| 24 |
+
"num_timestamp_prediction_blocks": 20
|
|
|
|
| 25 |
},
|
|
|
|
|
|
|
| 26 |
"initializer_range": 0.02,
|
| 27 |
"model_type": "fun_asr_nano",
|
| 28 |
"text_config": {
|
|
|
|
| 88 |
"vocab_size": 151936
|
| 89 |
},
|
| 90 |
"tie_word_embeddings": true,
|
| 91 |
+
"transformers_version": "5.15.0.dev0"
|
| 92 |
}
|
generation_config.json
CHANGED
|
@@ -4,6 +4,6 @@
|
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
"output_attentions": false,
|
| 6 |
"output_hidden_states": false,
|
| 7 |
-
"transformers_version": "5.
|
| 8 |
"use_cache": true
|
| 9 |
}
|
|
|
|
| 4 |
"eos_token_id": 151645,
|
| 5 |
"output_attentions": false,
|
| 6 |
"output_hidden_states": false,
|
| 7 |
+
"transformers_version": "5.15.0.dev0",
|
| 8 |
"use_cache": true
|
| 9 |
}
|
model.safetensors
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 sha256:335ca3e74917f1156690400e2c344350112950165789cf78ce3d0a367affd821
|
| 3 |
+
size 1659778184
|
processor_config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"audio_downsample_rate": 1,
|
| 3 |
"audio_token": "<|object_ref_start|>",
|
| 4 |
"default_transcription_prompt": "Transcribe the audio:",
|
| 5 |
"feature_extractor": {
|
|
@@ -15,6 +14,5 @@
|
|
| 15 |
"return_attention_mask": true,
|
| 16 |
"sampling_rate": 16000
|
| 17 |
},
|
| 18 |
-
"max_audio_len": null,
|
| 19 |
"processor_class": "FunAsrNanoProcessor"
|
| 20 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"audio_token": "<|object_ref_start|>",
|
| 3 |
"default_transcription_prompt": "Transcribe the audio:",
|
| 4 |
"feature_extractor": {
|
|
|
|
| 14 |
"return_attention_mask": true,
|
| 15 |
"sampling_rate": 16000
|
| 16 |
},
|
|
|
|
| 17 |
"processor_class": "FunAsrNanoProcessor"
|
| 18 |
}
|
tokenizer_config.json
CHANGED
|
@@ -21,7 +21,7 @@
|
|
| 21 |
"<|video_pad|>"
|
| 22 |
],
|
| 23 |
"is_local": true,
|
| 24 |
-
"local_files_only":
|
| 25 |
"model_max_length": 131072,
|
| 26 |
"pad_token": "<|endoftext|>",
|
| 27 |
"processor_class": "FunAsrNanoProcessor",
|
|
|
|
| 21 |
"<|video_pad|>"
|
| 22 |
],
|
| 23 |
"is_local": true,
|
| 24 |
+
"local_files_only": true,
|
| 25 |
"model_max_length": 131072,
|
| 26 |
"pad_token": "<|endoftext|>",
|
| 27 |
"processor_class": "FunAsrNanoProcessor",
|