Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Access Raw Audio after a Wake Word #9

Open
JpEncausse opened this issue Mar 29, 2024 · 6 comments
Open

Access Raw Audio after a Wake Word #9

JpEncausse opened this issue Mar 29, 2024 · 6 comments

Comments

@JpEncausse
Copy link

JpEncausse commented Mar 29, 2024

Following this discussion and my comment :

It would be very very valuable if we could have access to the raw audio after a wake word.
For instance : "Hello bot, (... some audio ...)"

  1. Because it open the possibility to call LLM like ChatGPT and play back answer
  2. Because offline trigger to get the audio is very smart compared to always listening that send tons of request.
  3. Or at least a sample code to record audio just after a wakeword (assuming we do not loss some part) ?

Many thanks for your help. I can test the GPT parts but looking for a sample to get started retrieveing the audio from microphone.

@qsjhyy
Copy link
Contributor

qsjhyy commented Apr 1, 2024

Unfortunately, this module does not support such functionality. Perhaps you can consider turning off the audio playback function of this module and adding a speech synthesis module to achieve this effect.

@JpEncausse
Copy link
Author

@qsjhyy For the speech synthesis I simply want to play an audio downloaded fron the cloud. It is the easy part because many online tools provide that audio. I do not need speech synthesis on board.

My issue is related on speech recognition. I want to get the audio buffer from the device after voice trigger. To send it to the cloud and perform Speech2Text online. The API do not provide access to the microphone buffer ?

@qsjhyy
Copy link
Contributor

qsjhyy commented Apr 2, 2024

The fact is, this module does not have the function that you say. This is more like the function of a standalone microphone.

@JpEncausse
Copy link
Author

Ahhhh ok, I though a software workaround would be possible (since it's a microphone) but I understand it is a hardware limitation.

That would be awesome if DFRobot that kind of hardware. Here is an exemple with Grok, no not the Elon LLM, not the CPU, but a plush that embed a microphone connected to the cloud.

@qsjhyy
Copy link
Contributor

qsjhyy commented Apr 8, 2024

DFRobot will release this kind of hardware in the future, so stay tuned.

@JpEncausse
Copy link
Author

Thanks !
And I assume I also can't play raw audio using the speaker ? I have to add another Speaker and Microphone to the ESP-32 If I want to add dynamic question/answer

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants