tts support (WIP)

[gilcu3]

This PR depends on #453. It adds support for the current tts models from openai. Feel free to try and let me know if anything breaks.

[rokipet]

Hello im in window i get this error how to fix it ? 2023-11-09 11:21:27,978 - root - ERROR - ⚠️ An error has occurred. ⚠️
[Errno 13] Permission denied: 'C:\Users\ADMINI1\AppData\Local\Temp\2\tmprutw7r3l'
Traceback (most recent call last):
File "C:\Users\Administrator\Downloads\Bot Updated\chatgpt-telegram-bot-
086f844\bot\openai_helper.py", line 366, in generate_speech
response.stream_to_file(temp_file.name)
File "C:\Program Files\Python311\Lib\site-packages\openai_base_client.py", line 1731, in stream_to_file
with open(file, mode="wb") as f:
^^^^^^^^^^^^^^^^^^^^^
PermissionError: [Errno 13] Permission denied: 'C:\Users\ADMINI1\AppData\Local\Temp\2\tmprutw7r3l'

Failed to generate speech: ⚠️ An error has occurred. ⚠️
Errno 13 Permission denied: 'C:\Users\ADMINI~1\AppData\Local\Temp\2\tmprutw7r3l'

[gilcu3]

Hello im in window i get this error how to fix it ? 2023-11-09 11:21:27,978 - root - ERROR - ⚠️ An error has occurred. ⚠️ [Errno 13] Permission denied: 'C:\Users\ADMINI1\AppData\Local\Temp\2\tmprutw7r3l' Traceback (most recent call last): File "C:\Users\Administrator\Downloads\Bot Updated\chatgpt-telegram-bot- 086f844\bot\openai_helper.py", line 366, in generate_speech response.stream_to_file(temp_file.name) File "C:\Program Files\Python311\Lib\site-packages\openai_base_client.py", line 1731, in stream_to_file with open(file, mode="wb") as f: ^^^^^^^^^^^^^^^^^^^^^ PermissionError: [Errno 13] Permission denied: 'C:\Users\ADMINI1\AppData\Local\Temp\2\tmprutw7r3l'

Failed to generate speech: ⚠️ An error has occurred. ⚠️ Errno 13 Permission denied: 'C:\Users\ADMINI~1\AppData\Local\Temp\2\tmprutw7r3l'

Ups, I did not know Windows would have issues with temporal files... I will change that to use files in memory, it will be more portable. Wait for my next commit on this branch.

[gilcu3]

@rokipet You can test now

[rokipet]

thanks that worked

[Gigabytedru]

I can't figure out how to use tts! is there an instruction? please

[gilcu3]

I can't figure out how to use tts! is there an instruction? please

If you cloned this branch, you can use it with /tts "any text to read"

[Gigabytedru]

I have a question, can I receive responses from the GPT chat in the format of MP3 in my Telegram bot?

[gilcu3]

I have a question, can I receive responses from the GPT chat in the format of MP3 in my Telegram bot?

It is technically possible, but not implemented yet. What's your use-case?

[kithawk]

I have a question, can I receive responses from the GPT chat in the format of MP3 in my Telegram bot?

It is technically possible, but not implemented yet. What's your use-case?

I can imagine a workflow when bot answers a query not with text but TTS voice message directly (so, even sending back and forth voice messages between the user and bot). Something like REPLY_WITH_TTS=true

[gilcu3]

I can imagine a workflow when bot answers a query not with text but TTS voice message directly (so, even sending back and forth voice messages between the user and bot). Something like REPLY_WITH_TTS=true

I see the point now, I thought it was about the format of the message, which is currently a voice message, but could also be just an mp3 file. Regarding adding support for REPLY_WITH_TTS=true I think it could be a good idea, if you can open an issue pointing that out so that it can get implemented later once this is merge and the code stabilizes.

[n3d1117]

Thanks @gilcu3! Tested this and seems to work great, made a couple of fixes and aligned all translations. Merging now! Feel free to open a separate PR for further improvements

[Jipok]

I can imagine a workflow when bot answers a query not with text but TTS voice message directly...

Like this #468 ?