no translation is still not working with gemini, i tried using api key off my another paid account. please !!!!! fix the issue..... 翻译仍然无法在Gemini上工作，我尝试使用我另一个付费账户的API密钥。请修复这个问题……

[abhijeet12s]

no translation is still not working with gemini, i tried using api key off my another paid account. it was working perfectly fine two days ago and suddenly its stopped working there is no issue with api key. i have checked everything translation still works on old version of pyvideotrans 1.13 , pyvideotrans 1.45 , pyvideotrans 1.75 but it clumps, combines and mixes two translated lines with each other . which why its harder to translate on those old versions. please !!!!! fix the issue , if you have any questions you can ask me.

issue : 翻译字幕C:/Users/abhij/YT Saver/Download/一口气看完韩剧《时间》女人为了调查妹妹死亡的真相选择嫁给了富豪2.srt出错:Gemini:2 retries after error persists :429 Resource has been exhausted (e.g. check quota).

没有翻译，Gemini仍然无法正常工作。我尝试使用另一个付费账户的API密钥。两天前它还工作得很好，但突然就停止了工作，API密钥没有问题。我已经检查了所有东西，翻译在旧版本的pyvideotrans 1.13、pyvideotrans 1.45和pyvideotrans 1.75上仍然有效，但它会将两行翻译混在一起，这使得在这些旧版本上翻译变得更困难。请！！！修复这个问题，如果你有任何问题，可以问我。

问题：翻译字幕C:/Users/abhij/YT Saver/Download/一口气看完韩剧《时间》女人为了调查妹妹死亡的真相选择嫁给了富豪2.srt出错:Gemini:2 retries after error persists:429 Resource has been exhausted (e.g. check quota).

[jianchang512]

https://groups.google.com/g/adwords-api/c/w7wKza2C_oI

https://www.googlecloudcommunity.com/gc/AI-ML/Gemini-API-429-Resource-has-been-exhausted-e-g-check-quota/m-p/728855

[jianchang512]

update to v2.22 patchUpdate

[abhijeet12s]

update to v2.22 patchUpdate

translation is still no working in v2.22 but its working in v1.75. see in the screenshots uploaded

[jianchang512]

[abhijeet12s]

still not working i have set translation_wait to 150 but it works on pyvideotrans 1.95 but in 1.95 trnslation is out of order can you merge gemini.py file in 2.23? please it will be helpfull !! why it works on 1.95 ? i think its uses gemini-pro model instead of gemini 1.5 pro thats why translation is working on pyvideotrans 1.95. you can check the code and screenshots below.

as you can see translation is working on 1.95 can you just swap gemini 1.5pro model to gemini-pro in 2.23 and send me sp.exe file

cose of gemini.py in 1.95 : # -- coding: utf-8 --

import re,os import time from videotrans.configure import config from videotrans.util import tools import google.generativeai as genai from google.generativeai.types import HarmCategory, HarmBlockThreshold

safetySettings = [ { "category": HarmCategory.HARM_CATEGORY_HARASSMENT, "threshold": HarmBlockThreshold.BLOCK_NONE, }, { "category": HarmCategory.HARM_CATEGORY_HATE_SPEECH, "threshold": HarmBlockThreshold.BLOCK_NONE, }, { "category": HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT, "threshold": HarmBlockThreshold.BLOCK_NONE, }, { "category": HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT, "threshold": HarmBlockThreshold.BLOCK_NONE, }, ]

shound_del=False def update_proxy(type='set'): global shound_del if type=='del' and shound_del: del os.environ['http_proxy'] del os.environ['https_proxy'] del os.environ['all_proxy'] shound_del=False elif type=='set': raw_proxy=os.environ.get('http_proxy') if not raw_proxy: proxy=tools.set_proxy() if proxy: shound_del=True os.environ['http_proxy'] = proxy os.environ['https_proxy'] = proxy os.environ['all_proxy'] = proxy

def get_error(num=5, type='error'): REASON_CN = { 2: "超出长度", 3: "安全限制", 4: "文字过度重复", 5: "其他原因" } REASON_EN = { 2: "The maximum number of tokens as specified", 3: "The candidate content was flagged for safety", 4: "The candidate content was flagged", 5: "Unknown reason" } forbid_cn = { 1: "被Gemini禁止翻译:出于安全考虑，提示已被屏蔽", 2: "被Gemini禁止翻译:由于未知原因，提示已被屏蔽" } forbid_en = { 1: "Translation banned by Gemini:for security reasons, the prompt has been blocked", 2: "Translation banned by Gemini:prompt has been blocked for unknown reasons" } if config.defaulelang == 'zh': return REASON_CN[num] if type == 'error' else forbid_cn[num] return REASON_EN[num] if type == 'error' else forbid_en[num]

def get_content(d,*,model=None,prompt=None): update_proxy(type='set') response=None try: message=prompt.replace('{text}',"\n".join(d)) response = model.generate_content( message ) config.logger.info(f'[Gemini]请求发送:{message=}')

    result = response.text.replace('##', '').strip().replace(''', '"').replace('"', "'")
    config.logger.info(f'[Gemini]返回:{result=}')
    if not result:
        raise Exception("fail")
    return result, response
except Exception as e:
    error=str(e)
    config.logger.error(f'[Gemini]请求失败:{error=}')
    if response and response.prompt_feedback.block_reason:
        raise Exception(get_error(response.prompt_feedback.block_reason, "forbid"))

    if error.find('User location is not supported') > -1 or error.find('time out') > -1:
        raise Exception("当前请求ip(或代理服务器)所在国家不在Gemini API允许范围")

    if response and len(response.candidates) > 0 and response.candidates[0].finish_reason not in [0, 1]:
        raise Exception(get_error(response.candidates[0].finish_reason))

    if response and len(response.candidates) > 0 and response.candidates[0].finish_reason == 1 and \
            response.candidates[0].content and response.candidates[0].content.parts:
        result = response.text.replace('##','').strip().replace(''', '"').replace('"', "'")
        return result,response
    raise Exception(error)

def trans(text_list, target_language="English", *, set_p=True, inst=None, stop=0, source_code="",is_test=False): """ text_list: 可能是多行字符串，也可能是格式化后的字幕对象数组 target_language: 目标语言 set_p: 是否实时输出日志，主界面中需要 """

try:
    genai.configure(api_key=config.params['gemini_key'])
    model = genai.GenerativeModel('gemini-pro', safety_settings=safetySettings)
except Exception as e:
    err = str(e)
    raise Exception(f'请正确设置http代理,{err}')

# 翻译后的文本
target_text = {"0": [],"srts":[]}
index = 0  # 当前循环需要开始的 i 数字,小于index的则跳过
iter_num = 0  # 当前循环次数，如果 大于 config.settings.retries 出错
err = ""
is_srt = False if  isinstance(text_list, str) else True
split_size = int(config.settings['trans_thread'])

prompt = config.params['gemini_template']
with open(config.rootdir+"/videotrans/gemini.txt",'r',encoding="utf-8") as f:
    prompt=f.read()
prompt=prompt.replace('{lang}', target_language)

# 切割为每次翻译多少行，值在 set.ini中设定，默认10
end_point="。" if config.defaulelang=='zh' else ' . '
# 整理待翻译的文字为 List[str]
if not is_srt:
    source_text = [t.strip() for t in text_list.strip().split("\n") if t.strip()]
else:
    source_text=[]
    for i,it in enumerate(text_list):
        source_text.append(it['text'].strip().replace('\n','.')+end_point)
split_source_text = [source_text[i:i + split_size] for i in range(0, len(source_text), split_size)]

while 1:
    if config.exit_soft or (config.current_status != 'ing' and config.box_trans != 'ing' and not is_test):
        return

    if iter_num >= config.settings['retries']:
        err=f'{iter_num}{"次重试后依然出错" if config.defaulelang == "zh" else " retries after error persists "}:{err}'
        break
    iter_num += 1

    if iter_num > 1:
        if set_p:
            tools.set_process(
                f"第{iter_num}次出错重试" if config.defaulelang == 'zh' else f'{iter_num} retries after error',btnkey=inst.init['btnkey'] if inst else "")
        time.sleep(10)

    response = None
    for i, it in enumerate(split_source_text):
        if config.exit_soft or (config.current_status != 'ing' and config.box_trans != 'ing' and not is_test):
            return
        if i < index:
            continue
        if stop > 0:
            time.sleep(stop)
        try:
            result,response=get_content(it,model=model,prompt=prompt)
            if inst and inst.precent < 75:
                inst.precent += 0.01
            if not is_srt:
                target_text["0"].append(result)
                if not set_p:
                    tools.set_process_box(text=result + "\n",func_name="fanyi",type="set")
                continue

            sep_res = tools.cleartext(result).split("\n")
            raw_len=len(it)
            sep_len=len(sep_res)
            # 如果返回数量和原始语言数量不一致，则重新切割
            if sep_len<raw_len:
                print(f'翻译前后数量不一致，需要重新切割')
                sep_res=tools.format_result(it,sep_res,target_lang=target_language)
            # if sep_len>raw_len:
            #     sep_res[raw_len-1]=".".join(sep_len[raw_len-1:])
            #     sep_res=sep_res[:raw_len]
            # if sep_len != raw_len:
            #     sep_res=[]
            #     for it_n in it:
            #         try:
            #             t,response=get_content([it_n.strip()],model=model,prompt=prompt)
            #         except Exception as e:
            #             config.logger.error(f'触发安全限制，{t=},{it_n=}')
            #             t="--"
            #         sep_res.append(t)

            for x, result_item in enumerate(sep_res):
                if x < len(it):
                    target_text["srts"].append(result_item.strip().rstrip(end_point))
                    if set_p:
                        tools.set_process(result_item + "\n", 'subtitle')
                        tools.set_process(config.transobj['starttrans'] + f' {i * split_size + x + 1} ',btnkey=inst.init['btnkey'] if inst else "")
                    else:
                        tools.set_process_box(text=result_item + "\n", func_name="fanyi",type="set")

            if len(sep_res) < len(it):
                tmp = ["" for x in range(len(it) - len(sep_res))]
                target_text["srts"] += tmp

        except Exception as e:
            err = str(e)
            break
        else:
            # 未出错
            err = ''
            iter_num = 0
            index = 0 if i <= 1 else i
    else:
        break
update_proxy(type='del')

if err:
    config.logger.error(f'[Gemini]翻译请求失败:{err=}')
    if err.lower().find("Connection error")>-1:
        err='连接失败 '+err
    raise Exception(f'Gemini:{err}')

if not is_srt:
    return "\n".join(target_text["0"])

if len(target_text['srts']) < len(text_list) / 2:
    raise Exception(f'Gemini:{config.transobj["fanyicuowu2"]}')

for i, it in enumerate(text_list):
    if i < len(target_text['srts']):
        text_list[i]['text'] = target_text['srts'][i]
    else:
        text_list[i]['text'] = ""
return text_list

[jianchang512]

I've tested it and it works fine at 60. You can try to deploy the source code and change the model to gemini-pro.

[abhijeet12s]

I've tested it and it works fine at 60. You can try to deploy the source code and change the model to gemini-pro.

but i dont know how to deploy source code ? i am using windows is there any tutorial on deploying source code ? after deploying the source code how can i open sp.exe interface ?

[jianchang512]

update to 2.24

[abhijeet12s]

update to 2.24

thanks now its translating properly.👍👍