element-android icon indicating copy to clipboard operation
element-android copied to clipboard

function to transcribe your voice notes to text.

Open discapacidad5 opened this issue 4 years ago • 8 comments

Your use case

function to transcribe your voice notes to text. Hello, here I am again with my requests for people with disabilities in this case for deaf and deafblind people. although I must admit that it is a function that also works for people without any disability since there are people who for some reason cannot listen to audio at that moment because they are in a meeting because they cannot disturb making noise or simply because the audio is very long and it is much faster to read than to listen. Whatever the reason, in short, audio-to-text transcription is definitely a very useful feature. In fact, I was looking at a news story that WhatsApp plans to incorporate this accessibility solution in the future and I wonder why we should be left behind with a function that is really not complicated since cell phones currently have the option of transcribing audio to text. , functions that can be incorporated into the keyboard and in many Android applications, why not incorporate it into the chat so that it transcribes voice notes?

Have you considered any alternatives?

an alternative can be a boot to which the voice memo was sent as a mention and returned it translated into text

Additional context

No response

discapacidad5 avatar Oct 14 '21 03:10 discapacidad5

Senders of voice messages should take care of what the recipient(s) will be able to understand. On some circumstances, sending text messages using the keyboard voice input is probably better than sending voice messages. Other arguments have been given by @SpiritCroc in SchildiChat/SchildiChat-android#95

It's also worth noting that in large public rooms, I have not seen a lot of voice messages, precisely because this is not suitable.

bmarty avatar Oct 19 '21 20:10 bmarty

I think that the public that currently uses the application does not use much audio but it does not happen if the public that is not currently using it is supposed to be an alternative for more people, so it must cover the needs, the more needs are covered, the greater will be the target audience With regard to the keyboard transcribing voice to text, it does so at the level of the person who speaks, but the problem is not the person who speaks, but the person who listens to it, it is not the same to transcribe what I speak to text than to transcribe an audio that they send me to text It is such an important utility that even WhatsApp itself will implement it in the next updates because it is not only useful for people with disabilities but also when someone is in a meeting and receives a voice note but you cannot play it, they can read it. silently there are many utilities that this function has

discapacidad5 avatar Oct 20 '21 18:10 discapacidad5

One that is more accurate, and supports more languages. https://huggingface.co/spaces/openai/whisper but it requires internet.

the other option is to use

https://developer.android.com/reference/android/speech/SpeechRecognizer

ghost avatar Nov 09 '22 08:11 ghost

Here's a video showing how this feature works on Telegram. Would love to see it here!

https://www.youtube.com/watch?v=JuodbGqjM50

Here's a tutorial of how someone set up a bot to do this: https://maya.land/monologues/2021/08/05/matrix-bot-transcribe-speech-audio-messages.html

Question: if we want this feature, is it something that first would need to be implemented on the Matrix server side?

RayBB avatar Dec 12 '22 21:12 RayBB

Here's a video showing how this feature works on Slack, which I think is much nicer than Telegram's! https://www.youtube.com/watch?v=K5gz9RqBs0k

RayBB avatar Jan 28 '23 20:01 RayBB

Any update on this feature? Really wish can have this feature

Stainless5792 avatar Mar 22 '23 03:03 Stainless5792

iOS 17 now does this in Messages—it's now a fairly mainstream expectation.

I don't have time to listen to minutes-long voice notes in convos I've missed out on, I just need to skim a transcription.

Understanding is seldom the problem, it's more one of privacy and appropriateness. I can't get my phone out and listen to voice notes in public if they might contain sensitive work information, nor would it be socially acceptable in a lot of circumstances versus skimming the contents and forwarding it to someone else to action if appropriate.

BTW, not an Apple user, but that's a nice feature.

HammyHavoc avatar Mar 27 '24 01:03 HammyHavoc

It has been on WhatsApp iOS for a while and seems to be coming to Android soon https://www.androidcentral.com/apps-software/whatsapp-testing-transcribe-voice-notes-feature

RayBB avatar Mar 27 '24 08:03 RayBB