File size: 1,877 Bytes
37b15c7
7cb0b8e
 
 
 
 
9dc25d9
7cb0b8e
 
 
 
 
 
 
9dc25d9
c4001d6
d4af723
 
c4001d6
37b15c7
7cb0b8e
 
 
 
 
 
 
 
 
 
 
 
 
 
d4af723
7cb0b8e
 
 
 
 
 
37b15c7
c4001d6
 
 
 
 
 
 
d4af723
c4001d6
d4af723
9895fa7
d4af723
c4001d6
37b15c7
7cb0b8e
 
9895fa7
 
 
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
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
from huggingface_hub import InferenceClient
import gradio as gr

client = InferenceClient(
    "mistralai/Mixtral-8x7B-Instruct-v0.1"
)

def format_prompt(message, history):
    prompt = "<s>"
    for user_prompt, bot_response in history:
        prompt += f"[INST] {user_prompt} [/INST]"
        prompt += f" {bot_response}</s> "
    prompt += f"[INST] {message} [/INST]"
    return prompt

def generate_from_srt(file_content, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
    # Process the SRT file content as needed before using it as a prompt
    # For example, extracting text and removing timestamps if necessary
    # Directly using the file content for simplicity here
    
    temperature = float(temperature)
    if temperature < 1e-2:
        temperature = 1e-2
    top_p = float(top_p)

    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=True,
        seed=42,
    )

    formatted_prompt = format_prompt(file_content, [])
    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
    output = ""

    for response in stream:
        output += response.token.text
    return output

def handle_file(file_info):
    # Directly use the file content if it's a text file
    if isinstance(file_info, str):
        file_content = file_info
    else:
        # If file_info is not a string, it might be a binary file
        file_content = file_info.decode('utf-8')
    
    return generate_from_srt(file_content)

iface = gr.Interface(
    fn=handle_file,
    inputs=gr.File(label="Upload SRT File", type="text"),  # Specify file type as text
    outputs="text",
    title="SRT File Translation",
    concurrency_limit=20,
)

iface.launch()