File size: 1,101 Bytes
85f7435
a876ea7
 
df42c7c
 
419d9dd
df42c7c
 
a876ea7
094f4df
7e196a6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
---
base_model:
- openai/whisper-large-v3-turbo
datasets:
- mozilla-foundation/common_voice_17_0
- google/fleurs
language:
- th
library_name: transformers
pipeline_tag: automatic-speech-recognition
---

```python
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
import torch
# MODEL_NAME = "whisper-large-v3-turbo-thai"
MODEL_NAME = "whis"

device = "cuda:0" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

model = AutoModelForSpeechSeq2Seq.from_pretrained(
    MODEL_NAME,
    torch_dtype=torch_dtype,
    # low_cpu_mem_usage=True,
    # use_safetensors=True,
)
model.to(device)
processor = AutoProcessor.from_pretrained(MODEL_NAME)

whisper = pipeline(
    "automatic-speech-recognition",
    model=model,
    tokenizer=processor.tokenizer,
    feature_extractor=processor.feature_extractor,
    max_new_tokens=128,
    torch_dtype=torch_dtype,
    device=device,
)

whisper("c.mp3",chunk_length_s=30,stride_length_s=5,batch_size=16,return_timestamps=True,generate_kwargs  = {"language":"<|th|>"})
```