chore: Added ru locale, added voximplant adapter, groq provider, switched to whisper in deepgram (multilang) #304
Conversation
…ched to whisper in deepgram (multilang) bunch of chore code, not serius
| @@ -42,7 +43,7 @@ | |||
| 'anyscale': LiteLLM, | |||
| 'custom': OpenAiLLM, | |||
| 'ola': OpenAiLLM, | |||
| 'groq': LiteLLM, | |||
| 'groq': GroqLLM, | |||
There was a problem hiding this comment.
any specific reason of not using litellm for groq?
There was a problem hiding this comment.
Hi, I'm sorry for the very dirty pr, I'll correct it and write in the comments
There was a problem hiding this comment.
any specific reason of not using litellm for groq?
in cases when user need to add more then 1 litellm driven provider,
real case example:
I could be wrong, but in our project, which is a fork of your repository, we need to use both groq and antrophic at the same time, from env variables
| @@ -0,0 +1,28 @@ | |||
| aiobotocore==2.9.0 | |||
There was a problem hiding this comment.
these packages could be installed when installing bolna package in the docker
There was a problem hiding this comment.
I almost missed the mark, I'm not that good at python
| print(f'websocket_url: {websocket_url}') | ||
| print(f'recipient_phone_number: {recipient_phone_number}') | ||
|
|
||
| voxapi = VoximplantAPI("credentials/voximplant.json") |
There was a problem hiding this comment.
can you also please commit a test/temporary sample voximplant json file.
There was a problem hiding this comment.
Integration with Voximplant should be two-way, unfortunately there is no built-in ability to stream audio from a websocket, but on the bolna side everything is ok
bunch of chore code, not serius