# CloseCaptions/subtitules capturing 1 side audio

**URL:** <https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740>\
**Category:** Integration\
**Created:** [December 20, 2021, 1:18pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740 "2021-12-20T13:18:19Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![DaMarin94](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/damarin94/32/1368_2.png) [@DaMarin94](https://mediasoup.discourse.group/u/DaMarin94)\
**Post date:** [December 20, 2021, 1:18pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/1 "2021-12-20T13:18:19Z")

</div>

Hello, Im trying to add this feature to my proyect by consuming an external API where I can send audio (o microphone in real time) and it returns the string of the interpretation. My intention is to capture the audio from just 1 side of the videocall and show the text on the other side (for deaf people to use the service). I checked the documentation but Im not sure where should I add this code. Thanks and sorry if my question is out of place/category.

---

<div class="post-metadata">

**Author:** ![nazar-pc](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/nazar-pc/32/369_2.png) [@nazar-pc](https://mediasoup.discourse.group/u/nazar-pc)\
**Post date:** [December 20, 2021, 2:52pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/2 "2021-12-20T14:52:25Z")

</div>

mediasoup is a library, you don’t just add a piece of code somewhere, you need to build an app using it that does what you need. In This case you’ll have to consume audio from a particular speaker and send it somehow to the service that will do text recognition.

---

<div class="post-metadata">

**Author:** ![DaMarin94](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/damarin94/32/1368_2.png) [@DaMarin94](https://mediasoup.discourse.group/u/DaMarin94)\
**Post date:** [December 20, 2021, 3:01pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/3 "2021-12-20T15:01:21Z")

</div>

Thanks very much, I needed orientation on how plausible was what I was trying to do.

---

<div class="post-metadata">

**Author:** ![nazar-pc](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/nazar-pc/32/369_2.png) [@nazar-pc](https://mediasoup.discourse.group/u/nazar-pc)\
**Post date:** [December 20, 2021, 3:18pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/4 "2021-12-20T15:18:51Z")

</div>

This is nothing unusual, I think recognition services even support RTP in some cases, so you can catch RTP packets in Node.js and send to the service.

---

<div class="post-metadata">

**Author:** ![jbaudanza](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/jbaudanza/32/1996_2.png) [@jbaudanza](https://mediasoup.discourse.group/u/jbaudanza)\
**Post date:** [December 20, 2021, 4:26pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/5 "2021-12-20T16:26:52Z")

</div>

It’s definitely plausible. I’ve implemented something similar using ffmpeg and Google’s speech-to-text API.

---

<div class="post-metadata">

**Author:** ![DaMarin94](https://yyz2.discourse-cdn.com/free1/user_avatar/mediasoup.discourse.group/damarin94/32/1368_2.png) [@DaMarin94](https://mediasoup.discourse.group/u/DaMarin94)\
**Post date:** [December 20, 2021, 5:46pm UTC](https://mediasoup.discourse.group/t/closecaptions-subtitules-capturing-1-side-audio/3740/6 "2021-12-20T17:46:22Z")

</div>

Thanks both, Im going that way now that I know it’s posible.