v0.22.0

What's Changed

• fix: Few typos in Tortoise docs. by @VladCuciureanu in #3352
• fix pause problem of Chinese speech by @aaron-lii in #3351
• Fix typos by @omahs in #3368
• Print message for either commercial license or CPML by @JRMeyer in #3381
• Add inference parameters by @WeberJulian in #3373
• Training fastspeech2 with External Speaker Embeddings by @freds0 in #3404
• fixes a typo by @joelhoward0 in #3392
• support multiple GPU training for XTTS by @aaron-lii in #3391
• Add studio speakers to open source XTTS! by @WeberJulian in #3405

New Contributors

• @VladCuciureanu made their first contribution in #3352
• @aaron-lii made their first contribution in #3351
• @omahs made their first contribution in #3368
• @JRMeyer made their first contribution in #3381
• @freds0 made their first contribution in #3404
• @joelhoward0 made their first contribution in #3392

Full Changelog: v0.21.3...v0.22.0

v0.21.3

What's Changed

• Add XTTS Fine tuning gradio demo by @Edresson in #3296

Full Changelog: v0.21.2...v0.21.3

No-Code XTTS fine-tuning

We created a UI that you can use to fine-tune XTTS with your data. You can run it on Colab, locally, or on a server.

@WeberJulian has also recorded a video for showing step-by-step tutorial

You can also follow the XTTS docs if you are a read-and-learn type.

v0.21.2

What's Changed

• Run XTTS models by direct name with versions by @erogol in #3318
• fix: correctly strip/restore initial punctuation by @eginhard in #3336
• Fix link to installation instructions by @Vuizur in #3329

New Contributors

• @Vuizur made their first contribution in #3329

Full Changelog: v0.21.1...v0.21.2

This PR allows for running XTTS models with version tags. So you the user can access any version they like.

from TTS.api import TTS

# get v2.0.2
tts = TTS(model_name="xtts_v2.0.2", gpu=True)

# get the latest version
tts = TTS(model_name="xtts", gpu=True)

# generate speech by cloning a voice using default settings
tts.tts_to_file(text="Here is my sample text.",
                file_path="output.wav",
                speaker_wav=["reference.wav", "reference1.wav"],
                language="en")

Making automatic sentence splitting optional. So you can apply any custom logic for processing the text before passing it to the model. Set split_sentences False.

from TTS.api import TTS

# get v2.0.2
tts = TTS(model_name="xtts_v2.0.2", gpu=True)

# generate speech by cloning a voice using default settings
tts.tts_to_file(text="Here is my sample text.",
                file_path="output.wav",
                speaker_wav=["reference.wav", "reference1.wav"],
                language="en",
                split_sentences=False)

v0.21.1

• Adding a basic Hindi text cleaner. 3206513

v0.21.0

What's Changed

• Remove duplicate/unused code by @eginhard in #3243
• Making the Model Manager's Progress bar statically accessible via the class. by @FlorianEagox in #3297
• More informative error for wrong --language argument by @eginhard in #3294
• Don't pass quotes to espeak by @eginhard in #3286
• Fix tts_with_vc by @eginhard in #3275
• Misjudgment of is_multi_lingual When Loading Multilingual Model via model_path by @TITC in #3273
• Introducing Development Dockerfile by @Kaszanas in #3263
• update deepspeed version by @WeberJulian in #3281

New Contributors

• @FlorianEagox made their first contribution in #3297
• @TITC made their first contribution in #3273
• @Kaszanas made their first contribution in #3263

Full Changelog: v0.20.6...v0.21.0

v0.20.6

What's Changed

• Remove duplicate AudioProcessor code, fix ExtractTTSpectrogram.ipynb by @eginhard in #3230
• Add sentence splitting by @WeberJulian in #3227
• Fix zh bug by @WeberJulian in #3238
• Update versions by @erogol in #3248
• Ensures that only GPT model is in training mode during XTTS GPT training by @Edresson in #3241
• Loosen dependencies and make k_diffusion optional by @erogol in #3249
• Update XTTS v2.0.2 by @erogol in #3249

Full Changelog: v0.20.5...v0.20.6

v0.20.5

What's Changed

• Add speed control for inference by @WeberJulian in #3214
• Update README.md by @eltociear in #3215
• Fix XTTS GPT padding and inference issues by @Edresson in #3216

Full Changelog: v0.20.4...v0.20.5

v0.20.4

What's Changed

• Update XTTS cloning by @erogol in #3207
• fix max generation length for XTTS by @WeberJulian in #3208

Full Changelog: v0.20.3...v0.20.4

v0.20.3

What's Changed

• XTTS- Torchaudio should use proper backend to load audio by @gorkemgoknar in #3179
• PyTorch 2.1 Updates (Weight Norm and TorchAudio I/O) by @MattyB95 in #3176
• xtts/tokenizer: merge duplicate implementations of preprocess_text by @akx in #3170
• fix(formatters): set missing root_path attribute by @eginhard in #3182

Full Changelog: v0.20.2...v0.20.3

v0.20.2

What's Changed

• Add char limit warn by @WeberJulian in #3130
• Fix coqui api by @erogol in #3168
• Fix #3153 by @erogol in #3169
• Move FreeVCConfig to TTS.vc.configs (like all other config classes) by @akx in #3126
• Fix ModelManager.list_models() by @eginhard in #3128
• Fix for exception on streaming on last chunk by @gorkemgoknar in #3160
• Add lang code in XTTS doc by @WeberJulian in #3158
• Remove v1 doc and tests by @WeberJulian in #3172

New Contributors

• @eginhard made their first contribution in #3128
• @gorkemgoknar made their first contribution in #3160

Full Changelog: v0.20.1...v0.20.2