Instructions to use Hishambarakat/Bahraini_ASR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Hishambarakat/Bahraini_ASR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Hishambarakat/Bahraini_ASR")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Hishambarakat/Bahraini_ASR", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload bahraini_asr_codeswitching/tokenizer_config.json with huggingface_hub
Browse files
bahraini_asr_codeswitching/tokenizer_config.json
CHANGED
|
@@ -5,7 +5,18 @@
|
|
| 5 |
"clean_up_tokenization_spaces": true,
|
| 6 |
"eos_token": "<|endoftext|>",
|
| 7 |
"errors": "replace",
|
| 8 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"<|endoftext|>",
|
| 10 |
"<|startoftranscript|>",
|
| 11 |
"<|en|>",
|
|
@@ -113,16 +124,5 @@
|
|
| 113 |
"<|startofprev|>",
|
| 114 |
"<|nocaptions|>",
|
| 115 |
"<|notimestamps|>"
|
| 116 |
-
]
|
| 117 |
-
"is_local": false,
|
| 118 |
-
"language": "ar",
|
| 119 |
-
"local_files_only": false,
|
| 120 |
-
"model_max_length": 1024,
|
| 121 |
-
"pad_token": "<|endoftext|>",
|
| 122 |
-
"predict_timestamps": false,
|
| 123 |
-
"processor_class": "WhisperProcessor",
|
| 124 |
-
"return_attention_mask": false,
|
| 125 |
-
"task": "transcribe",
|
| 126 |
-
"tokenizer_class": "WhisperTokenizer",
|
| 127 |
-
"unk_token": "<|endoftext|>"
|
| 128 |
}
|
|
|
|
| 5 |
"clean_up_tokenization_spaces": true,
|
| 6 |
"eos_token": "<|endoftext|>",
|
| 7 |
"errors": "replace",
|
| 8 |
+
"is_local": false,
|
| 9 |
+
"language": "ar",
|
| 10 |
+
"local_files_only": false,
|
| 11 |
+
"model_max_length": 1024,
|
| 12 |
+
"pad_token": "<|endoftext|>",
|
| 13 |
+
"predict_timestamps": false,
|
| 14 |
+
"processor_class": "WhisperProcessor",
|
| 15 |
+
"return_attention_mask": false,
|
| 16 |
+
"task": "transcribe",
|
| 17 |
+
"tokenizer_class": "WhisperTokenizer",
|
| 18 |
+
"unk_token": "<|endoftext|>",
|
| 19 |
+
"additional_special_tokens": [
|
| 20 |
"<|endoftext|>",
|
| 21 |
"<|startoftranscript|>",
|
| 22 |
"<|en|>",
|
|
|
|
| 124 |
"<|startofprev|>",
|
| 125 |
"<|nocaptions|>",
|
| 126 |
"<|notimestamps|>"
|
| 127 |
+
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
}
|