__version__ = (2, 0, 1) # █ █ ▀ █▄▀ ▄▀█ █▀█ ▀ # █▀█ █ █ █ █▀█ █▀▄ █ # © Copyright 2022 # https://t.me/hikariatama # # 🔒 Licensed under the GNU AGPLv3 # 🌐 https://www.gnu.org/licenses/agpl-3.0.html # meta pic: https://static.dan.tatar/vtt_icon.png # meta banner: https://mods.hikariatama.ru/badges/vtt.jpg # meta developer: @hikarimods # scope: ffmpeg # scope: hikka_only # scope: hikka_min 1.3.3 # requires: pydub speechrecognition python-ffmpeg import asyncio import logging import os import tempfile import speech_recognition as sr from pydub import AudioSegment from telethon.tl.types import DocumentAttributeVideo, Message from .. import loader, utils logger = logging.getLogger(__name__) @loader.tds class VoicyMod(loader.Module): """Recognize voice messages, audios, videos and round messages""" strings = { "name": "Voicy", "converting": ( "🫠 Recognizing voice" " message..." ), "converted": ( "🫠" " Recognized:\n{}" ), "voice_not_found": ( "🫠 Voice not found" ), "autovoice_off": ( "🫠 I will not recognize" " voice messages in this chat" ), "autovoice_on": ( "🫠 I will recognize" " voice messages in this chat" ), "_cfg_lang": "Language of voices to recognize", "_cfg_engine": "Recognition engine", "error": "🚫 Recognition error!", "_cfg_ignore_users": "Users to ignore", "_cfg_silent": "Silent mode - do not notify about errors", "too_big": "🫥 Voice message is too big, I can't recognise it...", } strings_ru = { "converting": ( "🫠 Распознаю голосовое" " сообщение..." ), "converted": ( "🫠" " Распознано:\n{}" ), "voice_not_found": ( "🫠 Нет ответа на" " войс" ), "autovoice_off": ( "🫠 Я больше не буду" " распознавать голосовые сообщения в этом чате" ), "autovoice_on": ( "🫠 Я буду распознавать" " голосовые сообщения в этом чате" ), "_cmd_doc_voicy": "Распознает голосовое сообщение", "_cmd_doc_autovoice": ( "Включить\\выключить автораспознавание голосовых сообщений в чате" ), "_cls_doc": "Распознает голосовые сообщения, аудио, видео и кругляши", "_cfg_lang": "Язык для распознавания голосовых сообщений", "_cfg_engine": "Распознаватель", "_cfg_ignore_users": "Игнорировать пользователей", "_cfg_silent": "Тихий режим - не оповещать об ошибках", "error": "🚫 Ошибка распознавания!", "too_big": ( "🫥 Голосовое сообщение слишком большое, я не могу его распознать..." ), } strings_de = { "converting": ( "🫠 Sprachnachricht wird" " erkannt..." ), "converted": ( "🫠" " Erkannt:\n{}" ), "voice_not_found": ( "🫠 Keine Antwort auf" " Voice" ), "autovoice_off": ( "🫠 Ich werde in diesem" " Chat keine Sprachnachrichten mehr erkennen" ), "autovoice_on": ( "🫠 Ich werde in diesem" " Chat Sprachnachrichten erkennen" ), "_cmd_doc_voicy": "Erkennt eine Sprachnachricht", "_cmd_doc_autovoice": ( "Aktiviert\\Deaktiviert die automatische Erkennung von Sprachnachrichten im" " Chat" ), "_cls_doc": "Erkennt Sprachnachrichten, Audios, Videos und Rundnachrichten", "_cfg_lang": "Sprache für die Spracherkennung", "_cfg_engine": "Erkennungsprogramm", "_cfg_ignore_users": "Benutzer ignorieren", "_cfg_silent": "Stiller Modus - Fehler nicht melden", "error": "🚫 Erkennungsfehler!", "too_big": ( "🫥 Sprachnachricht ist zu groß, ich kann sie nicht erkennen..." ), } strings_tr = { "converting": ( "🫠 Sesli mesajı" " tanıyorum..." ), "converted": ( "🫠" " Tanımlandı:\n{}" ), "voice_not_found": ( "🫠 Sesli mesaja cevap" " yok" ), "autovoice_off": ( "🫠 Bu sohbetteki sesli" " mesajları artık tanımayacağım" ), "autovoice_on": ( "🫠 Bu sohbetteki sesli" " mesajları tanıyacağım" ), "_cmd_doc_voicy": "Sesli mesajı tanır", "_cmd_doc_autovoice": ( "Sohbetteki sesli mesajların otomatik tanınmasını etkinleştirir\\devre dışı" " bırakır" ), "_cls_doc": "Sesli mesajları, sesleri, videoları ve çevirileri tanır", "_cfg_lang": "Ses tanıma için dil", "_cfg_engine": "Tanıyıcı", "_cfg_ignore_users": "Kullanıcıları yoksay", "_cfg_silent": "Sessiz mod - hataları bildirmeyin", "error": "🚫 Tanıma hatası!", "too_big": "🫥 Sesli mesaj çok büyük, tanıyamıyorum...", } strings_uz = { "converting": ( "🫠 So'zli xabar" " aniqlanmoqda..." ), "converted": ( "🫠" " Aniqlandi:\n{}" ), "voice_not_found": ( "🫠 So'zli xabarga" " javob yo'q" ), "autovoice_off": ( "🫠 Bu suhbatda so'zli" " xabarlar aniqlanmaydi" ), "autovoice_on": ( "🫠 Bu suhbatda so'zli" " xabarlar aniqlanadi" ), "_cmd_doc_voicy": "So'zli xabarni aniqlash", "_cmd_doc_autovoice": ( "Suhbatdagi so'zli xabarlar avtomatik aniqlashini yoqish\\o'chirish" ), "_cls_doc": "So'zli xabarlar, audio, videolar va qarishmalarni aniqlaydi", "_cfg_lang": "Tilni aniqlash uchun", "_cfg_engine": "Aniqlash moliyaviyasi", "_cfg_ignore_users": "Foydalanuvchilarni e'tiborsiz qoldirish", "_cfg_silent": "Sessiz rejim - xatolarni bildirmang", "error": "🚫 Aniqlash xatosi!", "too_big": "🫥 So'zli xabar juda katta, aniqlay olmayman...", } strings_hi = { "converting": ( "🫠 वायस संदेश" " पहचान रहा है..." ), "converted": ( "🫠" " पहचान लिया:\n{}" ), "voice_not_found": ( "🫠 वायस संदेश" " के लिए जवाब नहीं" ), "autovoice_off": ( "🫠 इस चैट में वायस" " संदेश पहचान नहीं करेंगे" ), "autovoice_on": ( "🫠 इस चैट में वायस" " संदेश पहचान करेंगे" ), "_cmd_doc_voicy": "वायस संदेश पहचान करें", "_cmd_doc_autovoice": "इस चैट में वायस संदेशों को ऑटोमैटिक पहचानने को सक्षम\\अक्षम करें", "_cls_doc": "वायस संदेश, ऑडियो, वीडियो और रैडियो संदेश पहचानता है", "_cfg_lang": "पहचान के लिए भाषा", "_cfg_engine": "पहचानकर्ता", "_cfg_ignore_users": "उपयोगकर्ताओं को नजरअंदाज करें", "_cfg_silent": "शांत मोड - त्रुटियों को सूचित न करें", "error": "🚫 पहचान त्रुटि!", "too_big": "🫥 वायस संदेश बहुत बड़ा है, पहचान नहीं कर सकता...", } def __init__(self): self.config = loader.ModuleConfig( loader.ConfigValue( "language", "ru-RU", lambda: self.strings("_cfg_lang"), validator=loader.validators.RegExp(r"^[a-z]{2}-[A-Z]{2}$"), ), loader.ConfigValue( "ignore_users", [], lambda: self.strings("_cfg_ignore_users"), validator=loader.validators.Series( validator=loader.validators.TelegramID() ), ), loader.ConfigValue( "silent", False, lambda: self.strings("_cfg_silent"), validator=loader.validators.Boolean(), ), ) async def client_ready(self): self.v2a = await self.import_lib( "https://libs.hikariatama.ru/v2a.py", suspend_on_error=True, ) self.chats = self.pointer("chats", []) async def recognize(self, message: Message): try: m = await utils.answer(message, self.strings("converting")) with tempfile.TemporaryDirectory() as tmpdir: file = os.path.join( tmpdir, "audio.mp3" if message.audio else "audio.ogg", ) data = await message.download_media(bytes) if message.video: data = await self.v2a.convert(data, "audio.ogg") with open(file, "wb") as f: f.write(data) song = AudioSegment.from_file( file, format="mp3" if message.audio else "ogg" ) song.export(os.path.join(tmpdir, "audio.wav"), format="wav") r = sr.Recognizer() with sr.AudioFile(os.path.join(tmpdir, "audio.wav")) as source: audio_data = r.record(source) text = await utils.run_sync( r.recognize_google, audio_data, language=self.config["language"], ) m = await utils.answer( m, self.strings("converted").format(text), ) except Exception: logger.exception("Can't recognize") if not self.config["silent"]: m = await utils.answer(m, self.strings("error")) await asyncio.sleep(3) if not message.out: await m.delete() @loader.unrestricted async def voicycmd(self, message: Message): """Recognize voice message""" reply = await message.get_reply_message() try: is_voice = ( reply.video or reply.audio or reply.media.document.attributes[0].voice ) except (AttributeError, IndexError): is_voice = False if not reply or not reply.media or not is_voice: await utils.answer(message, self.strings("voice_not_found")) return if message.out: await message.delete() await self.recognize(reply) if message.out: await message.delete() async def watcher(self, message: Message): try: if ( utils.get_chat_id(message) not in self.get("chats", []) or not message.media or not message.video and not message.audio and not message.media.document.attributes[0].voice or message.gif or message.sticker ): return except Exception: return if message.sender_id in self.config["ignore_users"]: return if ( ( message.video and ( next( attr for attr in message.video.attributes if isinstance(attr, DocumentAttributeVideo) ).duration > 120 ) ) or getattr( ( getattr( getattr(getattr(message, "media", None), "document", None), "attributes", False, ) or [None] )[0], "duration", 0, ) > 300 or message.document.size / 1024 / 1024 > 5 ): if not self.config["silent"]: await utils.answer(message, self.strings("too_big")) return await self.recognize(message) async def autovoicecmd(self, message: Message): """Toggle automatic recognition in current chat""" chat_id = utils.get_chat_id(message) if chat_id in self.get("chats", []): self.chats.remove(chat_id) await utils.answer(message, self.strings("autovoice_off")) else: self.chats.append(chat_id) await utils.answer(message, self.strings("autovoice_on"))