Ask: Are there any parameters that can be used to speed up translation? And how does it work?

[JinTTTT]

my example: A 600KB .epub file takes 30 minutes to translate using GPT3.5

[yihong0618]

sorry we do not use concurrency at first, only you can do is choose a faster model like groq

[mkXultra]

Perhaps using -batch might make it faster.

bilingual_book_maker/book_maker/cli.py

Lines 315 to 326 in a36b488

	parser.add_argument(
	"--batch",
	dest="batch_flag",
	action="store_true",
	help="Enable batch translation using ChatGPT's batch API for improved efficiency",
	)
	parser.add_argument(
	"--batch-use",
	dest="batch_use_flag",
	action="store_true",
	help="Use pre-generated batch translations to create files. Run with --batch first before using this option",
	)

Because it is a batch run, there is no rate limit. However, according to openai, the batch execution will be completed within 24 hours.

[mkXultra]

I also recommend the gpt4o-mini model.
It is faster, cheaper and better quality.

[uyiewnil]

I also recommend the gpt4o-mini model. It is faster, cheaper and better quality.

@mkXultra How can I use the 4o-mini model? I've tried a few methods, but none of them seem to work.

[mkXultra]

@uyiewnil
like this

python3 make_book.py --book_name ./Clarimonde.epub \
                                               --openai_key {api_key}  \
                                               --translate-tags h1,h2,h3,div,p \
                                               --model gpt4omini --language ja

if you want to use batch

 python3 make_book.py --book_name ./Clarimonde.epub \
                                                   --openai_key {api_key}  \
                                                   --translate-tags h1,h2,h3,div,p \
                                                   --language ja \
                                                   --model gpt4omini \
                                                   --batch
// few min later
 python3 make_book.py --book_name ./Clarimonde.epub \
                                                   --openai_key {api_key}  \
                                                   --translate-tags h1,h2,h3,div,p \
                                                   --language ja \
                                                   --model gpt4omini \
                                                   --batch-use

[JinTTTT]

@mkXultra
I tried using --batch, and afterward --batch-use, but it didn't work on my side, just keep telling me not cemplete yet

[mkXultra]

hmm, You can check the status of it in the section called batches in the open api.