{"id":3463,"date":"2023-06-05T14:57:29","date_gmt":"2023-06-05T05:57:29","guid":{"rendered":"https:\/\/blog.since2020.jp\/?p=3463"},"modified":"2023-06-05T14:59:10","modified_gmt":"2023-06-05T05:59:10","slug":"rpa_ai","status":"publish","type":"post","link":"https:\/\/since2020.jp\/media\/rpa_ai\/","title":{"rendered":"\u3010AI\u30c4\u30fc\u30eb\u6d3b\u7528\u3011\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u304b\u3089\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u81ea\u52d5\u3067\u4f5c\u6210\u3059\u308b\u30c4\u30fc\u30eb\u3092\u5b9f\u88c5\u3057\u3066\u307f\u305f"},"content":{"rendered":"\n<p>\u4eca\u56de\u3001\uff13\u3064\u306eAI\u30c4\u30fc\u30eb\u3092\u4f7f\u7528\u3057\u3066\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u304b\u3089\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210\u3059\u308bRPA\u30c4\u30fc\u30eb\u3092\u5b9f\u88c5\u3057\u307e\u3057\u305f\u3002\u5177\u4f53\u7684\u306b\u306f\u3001Whisper\u3067\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u306e\u97f3\u58f0\u3092\u30c6\u30ad\u30b9\u30c8\u5316\u3057\u3001ChatGPT\u3068DeepL\u3092\u4f7f\u3063\u3066\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210\u3057\u307e\u3057\u305f\u3002\u3053\u306e\u8a18\u4e8b\u3067\u306f\u3001\u5404\u30c4\u30fc\u30eb\u306e\u8a73\u3057\u3044\u4f7f\u3044\u65b9\u3084\u3001\u5b9f\u88c5\u6642\u306e\u5de5\u592b\u70b9\u306a\u3069\u7d39\u4ecb\u3057\u307e\u3059\u3002<\/p>\n\n\n<h2>\u52d5\u753b\u3092\u97f3\u58f0\u30d5\u30a1\u30a4\u30eb\u306b\u5909\u63db<\/h2>\n\u307e\u305a\u3001\u5bfe\u8c61\u306e\u52d5\u753b\u3092Python\u3092\u4f7f\u7528\u3057\u3066\u97f3\u58f0\u5909\u63db\u3057\u3066\u3044\u304d\u307e\u3059\u3002Whisper\u306f\u52d5\u753b\u3068\u97f3\u58f0\u4e21\u65b9\u306b\u5bfe\u5fdc\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u52d5\u753b\u3060\u3068\u6587\u5b57\u8d77\u3053\u3057\u3059\u308b\u51e6\u7406\u306b\u6642\u9593\u304c\u304b\u304b\u308b\u305f\u3081\u3001\u305d\u306e\u524d\u51e6\u7406\u3068\u3057\u3066Python\u3067\u52d5\u753b\u304b\u3089\u97f3\u58f0\u3092\u62bd\u51fa\u3057\u3066\u4f7f\u7528\u3057\u307e\u3059\u3002\u4eca\u56de\u306fwov\u30d5\u30a1\u30a4\u30eb\u3092mp3\u306b\u5909\u63db\u3057\u307e\u3057\u305f\u3002\u4e0b\u8a18\u304c\u4f7f\u7528\u3057\u305f\u30b3\u30fc\u30c9\u306b\u306a\u308a\u307e\u3059\u3002\r\n<div class=\"hcb_wrap\">\r\n<pre class=\"prism line-numbers lang-python\" data-lang=\"Python\"><code>import ffmpeg \r\n\r\n# \u52d5\u753b\u3092\u8aad\u307f\u8fbc\u307f \r\nstream = ffmpeg.input(\"target_video.mov\") \r\n\r\n# \u51fa\u529b\u5f62\u5f0f\u306e\u6307\u5b9a \r\nstream = ffmpeg.output(stream, \"target_sound.mp3\") \r\n\r\n# \u5b9f\u884c \r\nffmpeg.run(stream)\r\n<\/code><\/pre>\r\n<\/div>\n\n<h2>Whisper API\u3067\u97f3\u58f0\u3092\u6587\u5b57\u8d77\u3053\u3057<\/h2>\n\u6b21\u306b\u3001\u5909\u63db\u3057\u305f\u97f3\u58f0\u304b\u3089Whisper API\u3092\u4f7f\u3063\u3066\u6587\u5b57\u8d77\u3053\u3057\u3092\u3057\u3066\u3044\u304d\u307e\u3059\u3002\u4e0b\u8a18\u30b3\u30fc\u30c9\u3092\u4f7f\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u97f3\u58f0\u3092\u6587\u5b57\u8d77\u3053\u3057\u3057\u305f\u5185\u5bb9\u3092\u30c6\u30ad\u30b9\u30c8\u30d5\u30a1\u30a4\u30eb\u306b\u4fdd\u5b58\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3059\u3002\r\n<div class=\"hcb_wrap\">\r\n<pre class=\"prism line-numbers lang-python\" data-lang=\"Python\"><code>import openai \r\n\r\n# API\u30ad\u30fc\u306e\u8a2d\u5b9a \r\nopenai.api_key = '' \r\nopenai.organization = '' \r\n\r\naudio_file= open('target_audio.mp3', \"rb\") \r\ntranscript = openai.Audio.transcribe(\"whisper-1\", audio_file, language='ja', response_format='text') \r\nf = open('audio_to_text.txt', 'w') \r\nf.write(transcript) \r\nf.close()<\/code><\/pre>\r\n<\/div>\n\n<h2>\u30c6\u30ad\u30b9\u30c8\u3092\u5206\u5272\u3057\u3001DeepL\u3067\u82f1\u8a33<\/h2>\n\u6b21\u306b\u3001\u30c6\u30ad\u30b9\u30c8\u30d5\u30a1\u30a4\u30eb\u3092\u5206\u5272\u3057\u307e\u3059\u3002\u4eca\u56de\u306e\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u306f30\u5206\u3082\u3042\u308a\u3001\u6587\u5b57\u6570\u304c\u975e\u5e38\u306b\u591a\u304f\u306a\u308a\u307e\u3057\u305f\u3002ChatGPT\u306e\u30c8\u30fc\u30af\u30f3\u6570\u306e\u4e0a\u9650\u3092\u56de\u907f\u3059\u308b\u305f\u3081\u3001\u300c\u30a4\u30f3\u30c8\u30ed\u300d\u300c\u30e1\u30a4\u30f3\u300d\u300c\u30a2\u30a6\u30c8\u30ed\u300d\u306b\u76ee\u8996\u3067\u5206\u5272\u3057\u307e\u3057\u305f\u3002\u3053\u306e\u969b\u3001\u5185\u5bb9\u3092\u7406\u89e3\u3057\u3066\u5206\u5272\u305b\u305a\u3001\u30ad\u30fc\u30d5\u30ec\u30fc\u30ba\u3067\u5206\u5272\u57fa\u6e96\u3092\u5224\u65ad\u3057\u307e\u3057\u305f\u3002\u307e\u305f\u3001\u4eca\u56de\u306e\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u306e\u4f5c\u6210\u306b\u3042\u305f\u308a\u3001ChatGPT\u306b\u7d30\u304b\u3044\u6761\u4ef6\u3092\u8a2d\u5b9a\u3059\u308b\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u6295\u3052\u308b\u5fc5\u8981\u304c\u3042\u308a\u307e\u3057\u305f\u3002ChatGPT\u306f\u65e5\u672c\u8a9e\u3060\u3068\u6587\u5b57\u6570\u3067\u30c8\u30fc\u30af\u30f3\u6570\u306e\u7b97\u51fa\u3092\u884c\u3046\u306e\u306b\u5bfe\u3057\u3001\u82f1\u8a9e\u3060\u3068\u5358\u8a9e\u6570\u3067\u30c8\u30fc\u30af\u30f3\u6570\u306e\u7b97\u51fa\u3092\u884c\u3044\u307e\u3059\u3002\u305d\u306e\u305f\u3081\u3001\u540c\u3058\u5bb9\u91cf\u306e\u6587\u7ae0\u3067\u3082\u82f1\u8a9e\u306e\u65b9\u304c\u30c8\u30fc\u30af\u30f3\u6570\u3092\u7bc0\u7d04\u3067\u304d\u307e\u3059\u3002\u5f93\u3063\u3066\u3001\u4eca\u56de\u306f\u6587\u5b57\u8d77\u3053\u3057\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u3068ChatGPT\u306b\u6295\u3052\u308b\u30d7\u30ed\u30f3\u30d7\u30c8\u3092\u82f1\u8a33\u3059\u308b\u3053\u3068\u3067\u3001\u30c8\u30fc\u30af\u30f3\u6570\u306e\u7bc0\u7d04\u3092\u884c\u3044\u307e\u3057\u305f\u3002\n\n<h2>\u82f1\u8a33\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u3068\u30d7\u30ed\u30f3\u30d7\u30c8\u3092ChatGPT\uff08GPT-4\uff09\u3067\u5b9f\u884c<\/h2>\n\u4eca\u56de\u3001\u30c6\u30ad\u30b9\u30c8\u6587\u306e\u9577\u3055\u3068\u30d7\u30ed\u30f3\u30d7\u30c8\u306e\u7d30\u304b\u3055\u304b\u3089GPT-4\u3092\u4f7f\u7528\u3057\u307e\u3057\u305f\u3002\u4e0b\u8a18\u304c\u30d7\u30ed\u30f3\u30d7\u30c8\u306e\u4f8b\u3067\u3059\u3002\r\n<div class=\"hcb_wrap\">\r\n<pre class=\"prism line-numbers lang-plain\" data-lang=\"Plain Text\"><code># premise \r\nThis text is a transcription of an audio recording of a technical seminar. I would like to create event report about this seminar. This text is a document about a case study. # Order From this text, create a section corresponding to the introduction of the event report to be created, according to the following output rules. \r\n\r\n# Output Rule \r\n\u30fbThe first person in the event report must always be you \r\n\u30fbPlease create a explanatory text of level of detail that will allow those who were unable to attend this event to read this event report and have the same level of understanding as the attendees. \r\n\u30fbPlease describe the contents of this text in as much detail and specificity as possible. \r\n\u30fbThis is an event report to be created after the event, so please describe it in the past tense \r\n\u30fbPlease output this text in blocks by main points, with a set of headings and explanatory text. \r\n\u30fbThe explanatory text in the output must be least 1,000 words\r\n \u30fbOutput of intro or outro , such as preface and summary, is not needed. \r\n\u30fbPlease state only what is written in this sentence. In other words, do not describe your impressions, consideration or imagined opinions. \r\n\u30fbPlease be aware of the connection and context of each block description. Although the blocks are divided, we would like them to be in a series. \r\n\u30fbDescribe the background, assumptions, customer wishes, and challenges, and describe step-by-step, sequentially, how you responded to the challenges and what the results were.<\/code><\/pre>\r\n\u6700\u5f8c\u306b\u3001\u3053\u308c\u3089\u306e\u30d7\u30ed\u30f3\u30d7\u30c8\u306e\u5b9f\u884c\u7d50\u679c\u3092\u518d\u5ea6DeepL\u3092\u4f7f\u7528\u3057\u3066\u65e5\u672c\u8a9e\u7ffb\u8a33\u3057\u307e\u3057\u305f\u3002\u3053\u3046\u3059\u308b\u3053\u3068\u3067\u3001\u30c8\u30fc\u30af\u30f3\u7bc0\u7d04\u3057\u306a\u304c\u3089\u65e5\u672c\u8a9e\u306e\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210\u3057\u307e\u3057\u305f\u3002\r\n\r\n<\/div>\n\n<h2>\u307e\u3068\u3081<\/h2>\n\u4eca\u56de\u3001Whisper\u3001DeepL\u3001ChatGPT\u3092\u4f7f\u3063\u3066\u3001\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u304b\u3089\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210\u3059\u308bRPA\u30c4\u30fc\u30eb\u3092\u4f5c\u6210\u3057\u307e\u3057\u305f\u3002\u3053\u306e\u4ed6\u306b\u3082\u3001\u8b70\u4e8b\u9332\u3084\u8aac\u660e\u4f1a\u306a\u3069\u69d8\u3005\u306a\u30e6\u30fc\u30b9\u30b1\u30fc\u30b9\u3067\u306e\u5fdc\u7528\u304c\u53ef\u80fd\u3060\u3068\u8003\u3048\u3066\u3044\u307e\u3059\u3002\u7686\u69d8\u3082\u662f\u975e\u8a66\u3057\u3066\u307f\u3066\u304f\u3060\u3055\u3044","protected":false},"excerpt":{"rendered":"<p>\u4eca\u56de\u3001\uff13\u3064\u306eAI\u30c4\u30fc\u30eb\u3092\u4f7f\u7528\u3057\u3066\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u304b\u3089\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210\u3059\u308bRPA\u30c4\u30fc\u30eb\u3092\u5b9f\u88c5\u3057\u307e\u3057\u305f\u3002\u5177\u4f53\u7684\u306b\u306f\u3001Whisper\u3067\u30bb\u30df\u30ca\u30fc\u52d5\u753b\u306e\u97f3\u58f0\u3092\u30c6\u30ad\u30b9\u30c8\u5316\u3057\u3001ChatGPT\u3068DeepL\u3092\u4f7f\u3063\u3066\u30a4\u30d9\u30f3\u30c8\u30ec\u30dd\u30fc\u30c8\u3092\u4f5c\u6210 [&hellip;]<\/p>\n","protected":false},"author":85,"featured_media":3476,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","swell_btn_cv_data":"","footnotes":"","_wp_rev_ctl_limit":""},"categories":[1249],"tags":[96,145,448,460,459,347],"class_list":["post-3463","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-knowledge","tag-ai","tag-api","tag-chat-gpt","tag-deepl","tag-rpa","tag-whisper"],"_links":{"self":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/3463","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/users\/85"}],"replies":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/comments?post=3463"}],"version-history":[{"count":0,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/3463\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media\/3476"}],"wp:attachment":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media?parent=3463"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/categories?post=3463"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/tags?post=3463"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}