Last active
May 26, 2023 12:02
-
-
Save baydakovss/6d71c8d2ceb45a12769479c6807b19c1 to your computer and use it in GitHub Desktop.
Translate big text to target language in chat mode OpenAI with keeping context to better translation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import openai | |
import nltk | |
nltk.download('punkt') | |
openai.api_key = "XXXXXXXX" | |
def split_sentences(text, max_length): | |
""" | |
splitting text into chunks by sentences, not exceeding a specified maximum . | |
""" | |
tokenizer = nltk.data.load('tokenizers/punkt/english.pickle') | |
sentences = tokenizer.tokenize(text) | |
result = [] | |
current_chunk = "" | |
for sentence in sentences: | |
if len(current_chunk) + len(sentence) < max_length: | |
current_chunk += sentence + " " | |
else: | |
result.append(current_chunk.strip()) | |
current_chunk = sentence + " " | |
if current_chunk: | |
result.append(current_chunk.strip()) | |
return result | |
def translate_file(input_file, output_file, source_language, target_language, max_chunk_size): | |
with open(input_file, "r") as f: | |
text = f.read() | |
chunks = split_sentences(text, max_chunk_size) | |
with open(output_file, "w") as f: | |
for chunk in chunks: | |
chunk = chunk.replace('^M', '\n') | |
chunk = chunk.replace('\n', '\\n') | |
prompt = f"Translate the following from {source_language} to {target_language}: {chunk}" | |
print(prompt + "\n") | |
result = openai.ChatCompletion.create( | |
#engine="text-davinci-003", | |
engine="gpt-3.5-turbo", | |
#prompt=chunk, | |
prompt=prompt, | |
#max_tokens=2000, | |
max_tokens=1024, | |
temperature=0.5, | |
n=1, | |
stop=None, | |
#frequency_penalty=0, | |
#presence_penalty=0, | |
#timeout=60, | |
) | |
translation = result.choices[0].text.strip() | |
print(result.choices[0].text.strip()) | |
f.write(translation + "\n") | |
input_file = "input.txt" | |
output_file = "output.txt" | |
source_language = "English" | |
target_language = "Russian" | |
max_chunk_size = 500 | |
translate_file(input_file, output_file, source_language, target_language, max_chunk_size) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment