Tts Tacotron2 German

🚀 Text-to-Speech (TTS) with Tacotron2

This project uses Tacotron2 to perform text-to-speech (TTS). The model is trained on a custom German dataset with 12 days of voice data using SpeechBrain.

🚀 Quick Start

This project implements text-to-speech (TTS) using Tacotron2, trained on a custom German dataset with 12 days of voice data using SpeechBrain. The model was trained for 39 epochs (English SpeechBrain models are typically trained for 750 epochs), so there is room for improvement, and the model is likely to be updated soon. Fortunately, the HiFiGAN vocoder can be used independently of language.

📦 Installation

Install SpeechBrain.

pip install speechbrain

💻 Usage Examples

Basic Usage

import torchaudio
from speechbrain.pretrained import Tacotron2
from speechbrain.pretrained import HIFIGAN

# Intialize TTS (tacotron2) and Vocoder (HiFIGAN)
tacotron2 = Tacotron2.from_hparams(source="padmalcom/tts-tacotron2-german", savedir="tmpdir_tts")
hifi_gan = HIFIGAN.from_hparams(source="speechbrain/tts-hifigan-ljspeech", savedir="tmpdir_vocoder")

# Running the TTS
mel_output, mel_length, alignment = tacotron2.encode_text("Die Sonne schien den ganzen Tag.")

# Running Vocoder (spectrogram-to-waveform)
waveforms = hifi_gan.decode_batch(mel_output)

# Save the waverform
torchaudio.save('example_TTS.wav',waveforms.squeeze(1), 22050)

Advanced Usage

The above example demonstrates basic text-to-speech functionality. You can modify the input text to generate different speech outputs. Additionally, you can explore other parameters and methods provided by the Tacotron2 and HIFIGAN models to achieve more advanced usage, such as adjusting speech speed, pitch, etc.

📄 License

This project is licensed under the Apache-2.0 license.