flexthink

Add Text-to-Speech functionality (Tacotron only)

2

This pull request aims to add Text-to-Speech (TTS) functionality to the SpeechBrain project with the Tacotron model: Authors: * Georges Abous-Rjeili * Artem Ploujnikov

Add Text-To-Speech functionality (DeepVoice, Tacotron, WaveNet)

1

This pull request aims to add Text-to-Speech (TTS) functionality to the SpeechBrain project with the following models: - DeepVoice3 - Tacotron2 - WaveNet DeepVoice3 and WaveNet are based on the...

TTS Eval: Add TTS evaluation (MOS estimation)

2

## What does this PR do? Add TTS evaluation models trained on the SOMOS dataset There should be no breaking changes - [x] Did you read the [contributor guideline](https://speechbrain.readthedocs.io/en/latest/contributing.html)? -...

[Work in Progress] Curriculum sampler for LibriSpeech-like datasets

1

Work in progress: A dataset implementation that dynamically selects samples from LibriSpeech-compatible datasets with the specified number of words, using alignment data. For LibriSpeech, the LibriSpeech-Alignments dataset is required. https://zenodo.org/record/2619474#.Y3WeoS_73Eo

Fix recipe test errors in Latent Diffusion and pre-commit tests

1

## What does this PR do? Fixes # Before submitting - [x] Did you read the [contributor guideline](https://speechbrain.readthedocs.io/en/latest/contributing.html)? - [x] Did you make sure your **PR does only one thing**,...

This pull request contains a modified version of Tokotron, a basic sequence-to-sequence TTS

It appears that batch inference is not currently supported. If the batch size is anything other than 1, inference fails In models/vallex.py, `inference()`: ```python assert y.shape[0] == 1, y.shape ```...

## What does this PR do? Introduces a simple TTS architecture based on discrete speech representations from self-supervised models - [x] Did you read the [contributor guideline](https://speechbrain.readthedocs.io/en/latest/contributing.html)? - [x] Did...