# Can I access the utterance recording?
**URL:** https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276
**Category:** Feedback / Support
**Tags:** support
**Created:** [September 23, 2021, 12:48pm UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276 "2021-09-23T12:48:48Z")
**Posts on this page:** 8
**Page:** 1
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [September 23, 2021, 12:48pm UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/1 "2021-09-23T12:48:48Z")
Heya,
I’m writing a live translation skill that allows me to be converse in multiple languages using a portable rPi build.
The issue i’m facing is I put Mycroft into conversation mode, have a fallback skill that does the translation, now this works fine if the utterance is in same language Mycroft is expecting(english) but say then a french speaker talks, then the TTS obviously doesn’t work.
From within the fallback skill can I access the original recording that was used to generate the text?
Thanks
Marc
---
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [September 23, 2021, 1:12pm UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/2 "2021-09-23T13:12:00Z")
Hmm just seen [this](https://community.openconversational.ai/t/adjusting-recording-duration-for-get-response/11011/2).
I suppose I could just save the recordings using this method, pull, translate and then remove them! Unless there’s an easier way?
---
### Author:  [@gez-mycroft](https://community.openconversational.ai/u/gez-mycroft)
#### Post date: [September 24, 2021, 6:34am UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/3 "2021-09-24T06:34:52Z")
Hey Marc,
You can set a custom save path in your [mycroft config](https://mycroft-ai.gitbook.io/docs/using-mycroft-ai/customizations/config-manager):
```auto
"listener": {
"save_path": "/home/user/mycroft-saved-data",
"save_utterances": true
}
```
Very hacky, but if you grab the file with the latest timestamp you should theoretically get the last utterance.
I can’t think of a way of doing it properly without modifying mycroft-core or modifying an STT plugin
---
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [September 28, 2021, 9:50am UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/4 "2021-09-28T09:50:12Z")
Yeah awesome thanks a lot, that’s perfect - here’s another question(I can ask another question in the forum if you’d rather).
I want 1 Mycroft instance, now our internal tool is called Kane, I use it to transfer windows, data, displays across our iPhones, TV screens, laptops in the offices “Kane give me my laptop on this device” - kane has facial and voice recognition so he knows what “this” is because he knows where you’re standing.
Now if I’m on my mobile and I want to send a voice clip to “The” Kane not “A” Kane(installed on mobile) is there a way to do that also? Basically, can I submit an audio recording rather than using a mic as input?
Thanks
Marc
---
### Author:  [@gez-mycroft](https://community.openconversational.ai/u/gez-mycroft)
#### Post date: [September 30, 2021, 12:40am UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/5 "2021-09-30T00:40:32Z")
Not an easy way that I can think of off the top of my head…
Your options are probably to either hack the speech service in core to do what you want. Or pipe the audio stream from your recording and set that as the audio input source - essentially emulate a mic.
Others might have different ideas though?
---
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [September 30, 2021, 9:17am UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/6 "2021-09-30T09:17:54Z")
Hmm yeah I bet I could pretend to be the mic somehow, I’ll investigate that while we see if anyone has any other ideas, thanks a lot gez!
---
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [January 3, 2022, 6:02pm UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/7 "2022-01-03T18:02:46Z")
Bumping an older thread here but found it, will update in a while when I figure out the integration steps but -[https://voicemeeter.com/vban-talkie-use-cases/](https://voicemeeter.com/vban-talkie-use-cases/)
---
### Author:  [@Marc\_Fielding](https://community.openconversational.ai/u/Marc_Fielding)
#### Post date: [January 3, 2022, 6:03pm UTC](https://community.openconversational.ai/t/can-i-access-the-utterance-recording/11276/8 "2022-01-03T18:03:23Z")
Might not quite give the level of integration I’m after so we might need to implement our own VBAN emitter - but in short problem solved.