backmeupplz / voicy Goto Github PK
View Code? Open in Web Editor NEW@voicybot Telegram bot main repository
Home Page: https://voicybot.com
License: MIT License
@voicybot Telegram bot main repository
Home Page: https://voicybot.com
License: MIT License
Hi,
First of all, thanks for your work. I'm trying to self hosting it, so I clone the repo, created .env, launch yarn to download packages and then launch yarn start as you say in the docs gives me an error.
I explore the package json so I ended launching yarn develop to generate folders, the dist folder is generated and the app.js is launched. It says "Worker XXX started" but the bot dosn't answer to any messages, neither start nor help.
I've tried to add custom console.log on app.js, and I saw that code dosn't reach function runApp() {}, stops before.
I'm on CentOS 7 environment.
Any ideas? Thanks a lot.
отправляю ему файл и через телефон (TelegramX) и через web. Не хочет и все. Не ошибок не чего не пишет, просто не отвечает, хоть чем ругнулся бы.
Файл: mp3
размер: 5.48МБ
продолжительность: 2 мин 23 сек
hi. i want to suggest a config feature.
Something like a config for chat where you can set if the message will be send normally or the moment when the trascription is ready.
for example normally the message is something like "voice recognition has started" and then replaced with the actual text,
the problem i have with this is that when the notification arrives on the watch or band i can read only the placeholder since it can't read the canged notification.
what i was thinking is more something like the message will be send the moment it finished the transcription.
Thanks if this idea will be taken in consideration.
How exactly I need to specify wit.ai in .env file ?
I've forked bot, but my forked version has only English for wit.ai engine.
For Google - everything is ok, a huge list of languages.
Would be great if you could send file in from other messengers.
Бот упал и не отвечает
У Google Speech есть enhanced models, которые в теории должны работать лучше дефолтных. Есть ли возможность использовать их? Как я понимаю для этого нужно только пару доп. ключей в запрос прокинуть, сам на TS не пишу, попытался в коде найти где формируется запрос и не понял где нужно подправить :с
If I forward a voice message to another chat, voicy should add the name of the person, which had made the message, at the beginning of the transcription.
hi
i'm used wit.ai for recognition language persian but error
peech recognition is not supported for language: Persian
how do you used wai.ai for recognation for persian?
tnx
Hallo, I am trying to run this wonderful bot on my srever.
I installed everything, but "yarn start" seems to be missing.
I tried "yarn distribute" and this hasppens:
yarn run v1.22.19
$ node dist/app.js
Worker 47722 started
Done in 11.07s.
Can you help me?
TIA
режим инлайна доступен? можно как нить прикрутить для использования в режиме ответа в личных чатах?
This should be very useful for sharing voice messages from whatsapp
I've provided German
and English
in the WIT_LANGUAGES
env variable.
When I start a chat with the bot directly, the voice recognition works fine.
If I invite my bot into a (my) group, voicy prints a message, let me choose the language. I choose it, the engine and voice language but the voice recognition does not work. No response from the bot, nothing in the logs, even when I set to ENVIRONMENT=development
So that they can use google speech once again
Bot does not work anymore.
How to reproduce:
Create a group chat, add Vociy and another bot into the group. This another bot sometimes sends voice messages to the group. (I have this bot forwarding messages from elsewhere for me) However, Voicy does not act at all regarding the bot's voice messages.
I suppose Voicy's configured to response to non-bot group users' messages only, by default. I'd like to dive into the code and file a PR to enable it (and make it optional) by myself, if it takes time of yours.
In that case, some ballpark ideas of what-to-do in the code base would be very much appreciated. (New to TG bots and JS here)
Thanks. :)
Hi! Congratulations, I really like this bot.
I have a functionality suggestion. It would be nice to add a new mode for transcribing audios only when a group member cites the audio message with a command like \transcribe
, for instance. In this way, the bot would only be activated when requested.
Thanks for your work!
Hi
First of all, I'm sorry that I'm taking your time.
This is not really an issue with this bot, it's a question I had.
The thing is, this bot supports Persian STT but the problem is when I try wit.ai as an API for my project, this is the response I get:
{
"error" : "Speech recognition is not supported for language: Persian",
"code" : "bad-request"
}
I'm using POST method for speech using the following command in terminal:
curl -XPOST 'https://api.wit.ai/speech?v=20170307' -i -L -H "Authorization: Bearer $MYTOKEN" -H "Content-Type: audio/mpeg3" --data-binary "@sample.mp3"
I just replace $MYTOKEN with my real token that I got in the app.
so my question is, how did you do that?
Expressing yourself against the violence committed by the Russian Federation is a very noble thing to do, specially in a platform full of Putin supporters such as Telegram.
However, adding "Putin and his cronies kill civilians in the war in Ukraine #stopputin" to the end of messages is very confusing and unprofessional. It gives the impression that the audio says that, when in fact it does not. One may say it is literally "shoving words onto your mouth", I guess.
Telegram offers a variety of places for the developer to express their opinions, such as the "About the bot" section, descriptions, /start
command, etc. It may not have as much of an impact as adding links to the messages, but it would get the message sent across the platform without having to affect the main functionality of the bot, which is to caption audio messages. Politics should not affect applications, specially those who are important to accessibility (such as Voicy). It's simply unfair.
На бесплатном heroku инстансе падает по лимиту памяти в 512Mb почти сразу, после развёртывания приложения.
З.Ы. Спрашиваю как php'шник ))
Hallo, I cloned this repo on my arm64 Debian server.
The problem is that now I have this error:
TypeError: Cannot read property 'send' of undefined
at /root/voicy/dist/helpers/urlToText.js:57:16
at new Promise ()
at urlToText (/root/voicy/dist/helpers/urlToText.js:47:12)
at /root/voicy/dist/handlers/handleAudio.js:74:83
at Generator.next ()
at fulfilled (/root/voicy/dist/handlers/handleAudio.js:5:58)
at processTicksAndRejections (internal/process/task_queues.js:97:5) {
Any idea?
Thanks
Luca
There is the awesome voice recognition tool developed by Mozilla which may be very useful. I think Common Voice is the best open artificial intelligence project yet and DeepSpeech a wonderful resource. Is it possible to insert it among the options for the service to use in the Telegram Bot? Thanks a lot!
Some other useful repositories:
I ask you to remove political messages because I can't use the bot anymore for my dev team :(
add hebrew support
Actually, today I have 7 chat messages when I see it, and ~ 20 when administrators delete it
also in private messages
if you want, WE LOVE VOICY, ty for work, but stop spam msg pls)))
My russian voice message is 30 seconds..but voicy is only able to convert just a beginning of a new words. Using default engine
If someone posts audio,and then deletes the audio post, or Voicy is lagged and more than one user posts audio, other chat members can't see clearly who spoke which words, when Voicy reposts as text.
Can there be an option added to the bot that prefixes any speech-to-text with the speakers name and @:
Jane Doe [@jane123]:
[Actual text]
Hi!
Current workflow for switching language if far from optimal, and I switch a lot between languages.
I have feature request to make switching between few most often used languages more comfortable.
Here are some ideas , but this ticket is not limited to them, it's about making it more ergonomic, so here I go:
/l en
- to switch to english[/]
icon , to get not only /help
but also /language
/language
(and /l
) to display on a top as first positions "top 5 most used languages" and later alphabetic list of all the rest.@voicyenbot
@voicydebot
@voicyplbot
dedicated for languages and no need to switch ! :)Just few ideas that come to my mind. I would love to see it more ergonomic to switch between languages. Thank you!
https://github.com/backmeupplz/voicy/blob/main/src/helpers/startWebhook.ts
I just realized that the unlock is per-chat (and not per-user, as I think the donation message heavily suggests).
Now, premised I think the text should be made explicit regardless of anything, if you don't want to make big changes to the codebase I'd at least wish to transfer my purchase to my small company group chat rather than my private chat with voicybot.
The documentation should specify ffmpeg
as a dependency of the project. Or at least it should not fail silently if no ffmpeg installation is present.
Anyway, thanks for this amazing bot!
Dear all
I've set up my own copy of voicy but all I get as response to a voice message is 👮 I couldn't recognize that
.
❓ How can I debug my installation and configuration?
Wit.ai logs doesn't show any requests, so I think it has something to do with the installation or the wit.ai keymap.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.