Hugging Face's logo

Xerror
/
XTTS-v2_NOVA

Text-to-Speech
coqui
Model card Files Community
XTTS-v2_NOVA / README.md
Xerror's picture
Xerror
Upload 15 files
e511f5e verified
preview code
|
raw
history blame
5.78 kB
metadata
license: other
license_name: coqui-public-model-license
license_link: https://coqui.ai/cpml
library_name: coqui
pipeline_tag: text-to-speech
widget:
  - text: Once when I was six years old I saw a magnificent picture

ⓍTTS_v2 - CarliG Fine-Tuned Model

This repository hosts a fine-tuned version of the ⓍTTS model, utilizing 2 minutes of unique voice lines from AtheneLive's CarliG AI, the iconic GPT4 Chatbot who went viral after the release of gpt4 api. The voice lines were sourced from athenes live streams which can be found here: AtheneLive George Carlin & CarliG livestream

CarliG

Listen to a sample of the ⓍTTS_v2 - CarliG Fine-Tuned Model:

Here's a CarliG mp3 voice line clip from the training data:

Features

  • 🎙️ Voice Cloning: Realistic voice cloning with just a short audio clip.
  • 🌍 Multi-Lingual Support: Generates speech in 17 different languages while maintaining CarliG's distinct voice.
  • 😃 Emotion & Style Transfer: Captures the emotional tone and style of the original voice.
  • 🔄 Cross-Language Cloning: Maintains the unique voice characteristics across different languages.
  • 🎧 High-Quality Audio: Outputs at a 24kHz sampling rate for clear and high-fidelity audio.

Supported Languages

The model supports the following 17 languages: English (en), Spanish (es), French (fr), German (de), Italian (it), Portuguese (pt), Polish (pl), Turkish (tr), Russian (ru), Dutch (nl), Czech (cs), Arabic (ar), Chinese (zh-cn), Japanese (ja), Hungarian (hu), Korean (ko), and Hindi (hi).

Usage in Roll Cage

🤖💬 Boost your AI experience with this Ollama add-on! Enjoy real-time audio 🎙️ and text 🔍 chats, LaTeX rendering 📜, agent automations ⚙️, workflows 🔄, text-to-image 📝➡️🖼️, image-to-text 🖼️➡️🔤, image-to-video 🖼️➡️🎥 transformations. Fine-tune text 📝, voice 🗣️, and image 🖼️ gens. Includes Windows macro controls 🖥️ and DuckDuckGo search.

ollama_agent_roll_cage (OARC) is a completely local Python & CMD toolset add-on for the Ollama command line interface. The OARC toolset automates the creation of agents, giving the user more control over the likely output. It provides SYSTEM prompt templates for each ./Modelfile, allowing users to design and deploy custom agents quickly. Users can select which local model file is used in agent construction with the desired system prompt.

Why This Model for Roll Cage?

The CarliG fine-tuned model was designed for the Roll Cage chatbot to enhance user interaction with a familiar and beloved voice. By incorporating CarliG's distinctive speech patterns and tone, Roll Cage becomes more engaging and entertaining. The addition of multi-lingual support and emotion transfer ensures that the chatbot can communicate effectively and expressively across different languages and contexts, providing a more immersive experience for users.

CoquiTTS and Resources

License

This model is licensed under the Coqui Public Model License. Read more about the origin story of CPML here.

Contact

Join our 🐸Community on Discord and follow us on Twitter. For inquiries, email us at info@coqui.ai.

Using 🐸TTS API:

from TTS.api import TTS
tts = TTS(model_path="D:/CodingGit_StorageHDD/Ollama_Custom_Mods/ollama_agent_roll_cage/AgentFiles/Ignored_TTS/XTTS-v2_CarliG/", 
              config_path="D:/CodingGit_StorageHDD/Ollama_Custom_Mods/ollama_agent_roll_cage/AgentFiles/Ignored_TTS/XTTS-v2_CarliG/config.json", progress_bar=False, gpu=True).to(self.device)

# generate speech by cloning a voice using default settings
tts.tts_to_file(text="It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.",
                file_path="output.wav",
                speaker_wav="/path/to/target/speaker.wav",
                language="en")

Using 🐸TTS Command line:

 tts --model_name tts_models/multilingual/multi-dataset/xtts_v2 \
     --text "Bugün okula gitmek istemiyorum." \
     --speaker_wav /path/to/target/speaker.wav \
     --language_idx tr \
     --use_cuda true

Using the model directly:

from TTS.tts.configs.xtts_config import XttsConfig
from TTS.tts.models.xtts import Xtts

config = XttsConfig()
config.load_json("/path/to/xtts/config.json")
model = Xtts.init_from_config(config)
model.load_checkpoint(config, checkpoint_dir="/path/to/xtts/", eval=True)
model.cuda()

outputs = model.synthesize(
    "It took me quite a long time to develop a voice and now that I have it I am not going to be silent.",
    config,
    speaker_wav="/data/TTS-public/_refclips/3.wav",
    gpt_cond_len=3,
    language="en",
)