Speech synthesis (TTS) in low-resource languages by training from scratch with Fastpitch and fine-tuning with HifiGan

Automatic transcriber made with the Nvidia NeMo AI toolkit. Used to transcribe speech to text in real-time from any source. Requires CUDA capable GPU to run on the local machine, if setup using virtual audio cables can transcribe the audio that is being played in real-time without any other requirements.

The simplest & most comprehensible tutorial on speaker identification with NVIDIA's `Nemo`.

Module for russian speech recognition using NVIDIA Nemo.

Audio profanity detector desktop app developed with PyQt5 using NVidia-Nemo tech.

Implementation of a Kazakh Speech-to-Text Model using the NVIDIA NeMo toolkit for efficient transcription of spoken Kazakh speech into text.

PodcastProject Analytics Toolkit - Project that creates analytics various input data. Exported data is intended to be used in a PodcastProject website