Skip to content

mrmanna/Nvidia_Nemo_FastPitch_TTS_Example

A comprehensive guide is available in my Medium article:

How to Build a High-Quality Text-to-Speech (TTS) System Locally with Nvidia NeMo FastPitch

After running command:

poetry install

We have to update gcc to 12 to install cython and youtokentome which are required.

conda install -c conda-forge gcc_linux-64=12

pip install cython youtokentome

and then we can run the app like

poetry run start <yourpdfile.pdf> <youroutputfile.wav>

Caution:

This code includes minimal error handling. I encourage you to enhance it by addressing any issues you encounter during use.

Further if you want change pitch or speed, you can do that manipulating the spectrogram, happy coding.

About

How to Build a High-Quality Text-to-Speech (TTS) System Locally with Nvidia NeMo FastPitch

Topics

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages