Create README.md

README.md

@@ -0,0 +1,78 @@
+---
+library_name: transformers
+license: mit
+datasets:
+- NhutP/VSV-1100
+- mozilla-foundation/common_voice_14_0
+- AILAB-VNUHCM/vivos
+language:
+- vi
+metrics:
+- wer
+base_model:
+- openai/whisper-base
+---
+## Introduction
+- We release a new model for Vietnamese speech regconition task.
+- We fine-tuned [openai/whisper-base(https://huggingface.co/openai/whisper-base) on our new dataset [VSV-1100](https://huggingface.co/datasets/NhutP/VSV-1100).
+
+## Training data
+
+| [VSV-1100](https://huggingface.co/datasets/NhutP/VSV-1100) | T2S* | [CMV14-vi](https://huggingface.co/datasets/mozilla-foundation/common_voice_14_0) |[VIVOS](https://huggingface.co/datasets/AILAB-VNUHCM/vivos)| [VLSP2021](https://vlsp.org.vn/index.php/resources) | Total|
+|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|
+|   1100 hours  |   11  hours |   3.04 hours  |    13.94  hours| 180 hours | 1308 hours |
+
+\* We use a text-to-speech model to generate sentences containing words that do not appear in our dataset.
+
+## WER result
+| [CMV14-vi](https://huggingface.co/datasets/mozilla-foundation/common_voice_14_0) | [VIVOS](https://huggingface.co/datasets/AILAB-VNUHCM/vivos) | [VLSP2020-T1](https://vlsp.org.vn/index.php/resources) | [VLSP2020-T2](https://vlsp.org.vn/index.php/resources) | [VLSP2021-T1](https://vlsp.org.vn/index.php/resources) | [VLSP2021-T2](https://vlsp.org.vn/index.php/resources) |[Bud500](https://huggingface.co/datasets/linhtran92/viet_bud500) |
+|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|
+|13.74|7.94|16.02|45.55| 17.25 | 12.32 | 6.63 |
+
+
+## Usage
+### Inference
+```python
+from transformers import WhisperProcessor, WhisperForConditionalGeneration
+import librosa
+# load model and processor
+processor = WhisperProcessor.from_pretrained("NhutP/ViWhisper-base")
+model = WhisperForConditionalGeneration.from_pretrained("NhutP/ViWhisper-base")
+model.config.forced_decoder_ids = None
+
+# load a sample
+array, sampling_rate = librosa.load('path_to_audio', sr = 16000) # Load some audio sample
+input_features = processor(array, sampling_rate=sampling_rate, return_tensors="pt").input_features 
+# generate token ids
+predicted_ids = model.generate(input_features)
+# decode token ids to text
+transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+```
+### Use with pipeline
+```python
+from transformers import pipeline
+pipe = pipeline(
+    "automatic-speech-recognition",
+    model="NhutP/ViWhisper-base",
+    max_new_tokens=128,
+    chunk_length_s=30,
+    return_timestamps=False,
+    device= '...' # 'cpu' or 'cuda'
+) 
+output = pipe(path_to_audio_samplingrate_16000)['text']
+```
+
+## Citation
+
+```
+@misc{VSV-1100,
+    author = {Pham Quang Nhut and Duong Pham Hoang Anh and Nguyen Vinh Tiep},
+    title = {VSV-1100: Vietnamese social voice dataset},
+    url = {https://github.com/NhutP/VSV-1100},
+    year = {2024}
+}
+```
+
+Also, please give us a star on github: https://github.com/NhutP/ViWhisper if you find our project useful
+
+Contact me at: 22521061@gm.uit.edu.vn (Pham Quang Nhut)