feat: refactor get_tts, get_speech_to_score_dict to raise immediatly errors in case of empty text/language/audio inputs

feat: changed default sample rate to 24000 for both german and english tts models

chore: avoid binary content for aip_trainer/lambdas/__init__.py

feat: add text to speech (TTS)

ci: hugginface space, move from docker to gradio sdk v5.6.0, add missing packages.txt with ffmpeg, pre-requirements.txt with pip, update gradio app to properly format informations to frontend, update tests

feat: first working gradio fronted with refactored lambda_handler and tests

feat: use alternate version of init_jit_model to try avoiding PermissionError on HuggingFAce

ci: improve env variables

feat: refactor get_resulting_string to reduce its complexity, improve some return typing

feat: improve logs and comments

feat: add dockerfile and save yml silero model within system temp folders to support docker container execution

feat: support pytorch and torchaudio, update test, add requirements-dev.txt

test: update python backend test coverage, remove some unused classes and functions (see README.md)

feat: substitute sample table selection with direct input, update TestDataset tests

feat: frontend, add a table with all entries from language df and pick the and their API

feat: use structlog to handle logging, add some logs

refactor: remove use of utilsFileIO.py, update jquery@3.7.1 and bootstrap@5.3.3

refactor: organize project with tests in a package, start following suggestions from pycharm, sonarlint and snyk