How to do get Auto-GPT to speak?

[FredTheViking]

I am using the docker version of the Auto-GPT v0.3.0 Releason, and I have an account with Elevanlabs. I am confident I have set up the Elevenlabs API key correctly, as well as the Voice ids, but I am getting this error:

Welcome to Auto-GPT! run with '--help' for more information.
Exception in thread Thread-1 (speak):
Traceback (most recent call last):
File "/usr/local/lib/python3.10/threading.py", line 1016, in _bootstrap_inner
self.run()
File "/usr/local/lib/python3.10/threading.py", line 953, in run
self._target(*self._args, **self._kwargs)
File "/app/autogpt/speech/say.py", line 23, in speak
success = voice_engine.say(text, voice_index)
File "/app/autogpt/speech/base.py", line 33, in say
return self._speech(text, voice_index)
File "/app/autogpt/speech/eleven_labs.py", line 82, in _speech
playsound("speech.mpeg", True)
File "/usr/local/lib/python3.10/site-packages/playsound.py", line 91, in _playsoundNix
import gi
ModuleNotFoundError: No module named 'gi'

In the .env file, I have done this:

ELEVENLABS_API_KEY=a2MYKEYAPIKEYa2
ELEVENLABS_VOICE_1_ID=cDMYVOICEIDX
ELEVENLABS_VOICE_2_ID=cMYVOICE2A

Of course, replace the values used to protect my information, but just to give you an idea of what it looks like. Auto-GPT other works without the --speak options. Any thoughts on what's going on? Is it safe for me to provide a log file? I am using the following command to run:

docker-compose run --rm auto-gpt --gpt4only --speak

[ankerelite]

1. This worked for me (win):
   pip uninstall playsound
   pip install playsound==1.2.2

That installs a previous version of playsound which seems to work better with this.

Research unearthed that Speak function does not work with docker yet. I was able to get it to work great with environmentvar or venv. Hopefully saves you a headache :)

[FredTheViking]

Thank you for your response, and it was very helpful. I am new here, and it did not occur to me to research the issues section until you pointed it out, and many seem to be having this issue with docker. I am also newb to Docker, which means I will probably wait for the project to address this issue instead mucking with the docker.

[bibyts]

1. This worked for me (win):
   pip uninstall playsound
   pip install playsound==1.2.2

That installs a previous version of playsound which seems to work better with this.

Research unearthed that Speak function does not work with docker yet. I was able to get it to work great with environmentvar or venv. Hopefully saves you a headache :)

Thanks for the info. I have autogpt running on docker today and ran into the issue with eleven labs voice as well. What do you mean by: " I was able to get it to work great with environmentvar or venv"

[ankerelite]

I usedmkenv autogptthenworkon autogptOn Jun 6, 2023, at 2:16 PM, bibyts ***@***.***> wrote: This worked for me (win): pip uninstall playsound pip install playsound==1.2.2 That installs a previous version of playsound which seems to work better with this. Research unearthed that Speak function does not work with docker yet. I was able to get it to work great with environmentvar or venv. Hopefully saves you a headache :) Thanks for the info. I have autogpt running on docker today and ran into the issue with eleven labs voice as well. What do you mean by: " I was able to get it to work great with environmentvar or venv" —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: ***@***.***>