Improvement to the selection of requests per second in the patch sent to the TTS API

[Marcoxvid]

Problem:
Error in API response when trying to generate a podcast due to too many concurrent requests in the same bulk, currently batches of 5.

Suggestion:
Implement front-end selection of the maximum number of requests per second for the API to cover all types of contracted plans, such as elevenlabs, free, starter, etc.

[Marcoxvid]

In the podcast creator project it is mentioned that there is an environment parameter: TTS_BATCH_SIZE=3 but it does not work with the notebook and I also did not find any mention of this in the project source.

[lfnovo]

Hi @Marcoxvid , TTS_BATCH_SIZE is now available to be used.