pattaraearth commited on
Commit
e6ad90a
·
verified ·
1 Parent(s): 9619195

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +61 -3
README.md CHANGED
@@ -1,3 +1,61 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ language:
4
+ - th
5
+ - en
6
+ base_model:
7
+ - openai/whisper-medium
8
+ pipeline_tag: automatic-speech-recognition
9
+ library_name: transformers
10
+ metrics:
11
+ - wer
12
+ ---
13
+
14
+ # Pathumma Whisper Medium (Th)
15
+
16
+ ## Model Description
17
+ Additional information is needed
18
+
19
+ ## Quickstart
20
+ You can transcribe audio files using the [`pipeline`](https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.AutomaticSpeechRecognitionPipeline) class with the following code snippet:
21
+ ```python
22
+ import torch
23
+ from transformers import pipeline
24
+
25
+ device = "cuda" if torch.cuda.is_available() else "cpu"
26
+ torch_dtype = torch.bfloat16 if torch.cuda.is_available() else torch.float32
27
+
28
+ lang = "th"
29
+ task = "transcribe"
30
+
31
+ pipe = pipeline(
32
+ task="automatic-speech-recognition",
33
+ model="nectec/Pathumma-whisper-th-medium",
34
+ torch_dtype=torch_dtype,
35
+ device=device,
36
+ )
37
+ pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language=lang, task=task)
38
+
39
+ text = pipe("audio_path.wav")["text"]
40
+ print(text)
41
+ ```
42
+
43
+ ## Limitations and Future Work
44
+ Additional information is needed
45
+
46
+ ## Acknowledgements
47
+ We extend our appreciation to the research teams engaged in the creation of the open speech model, including AIResearch, BiodatLab, Looloo Technology, SCB 10X, and OpenAI. We would like to express our gratitude to Dr. Titipat Achakulwisut of BiodatLab for the evaluation pipeline. We express our gratitude to ThaiSC, or NSTDA Supercomputer Centre, for supplying the LANTA used for model training, fine-tuning, and evaluation.
48
+
49
+ ## Pathumma Audio Team
50
+ *Pattara Tipaksorn*, Wayupuk Sommuang, Oatsada Chatthong, *Kwanchiva Thangthai*
51
+
52
+ ## Citation
53
+ ```
54
+ @misc{tipaksorn2024PathummaWhisper,
55
+ title = { {Pathumma Whisper Medium (TH)} },
56
+ author = { Pattara Tipaksorn and Wayupuk Sommuang and Oatsada Chatthong and Kwanchiva Thangthai },
57
+ url = { https://huggingface.co/nectec/Pathumma-whisper-th-medium },
58
+ publisher = { Hugging Face },
59
+ year = { 2024 },
60
+ }
61
+ ```