Instructions to use rav009/VoxCPM2-lora-manbo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- VoxCPM
How to use rav009/VoxCPM2-lora-manbo with VoxCPM:
import soundfile as sf from voxcpm import VoxCPM model = VoxCPM.from_pretrained("rav009/VoxCPM2-lora-manbo") wav = model.generate( text="VoxCPM is an innovative end-to-end TTS model from ModelBest, designed to generate highly expressive speech.", prompt_wav_path=None, # optional: path to a prompt speech for voice cloning prompt_text=None, # optional: reference text cfg_value=2.0, # LM guidance on LocDiT, higher for better adherence to the prompt, but maybe worse inference_timesteps=10, # LocDiT inference timesteps, higher for better result, lower for fast speed normalize=True, # enable external TN tool denoise=True, # enable external Denoise tool retry_badcase=True, # enable retrying mode for some bad cases (unstoppable) retry_badcase_max_times=3, # maximum retrying times retry_badcase_ratio_threshold=6.0, # maximum length restriction for bad case detection (simple but effective), it could be adjusted for slow pace speech ) sf.write("output.wav", wav, 16000) print("saved: output.wav") - Notebooks
- Google Colab
- Kaggle
VoxCPM2-LoRA:曼波讲故事 音色适配器
English | 中文
📖 模型简介 (Model Description)
English:
This is a LoRA adapter fine-tuned on the VoxCPM2 base model. It is designed to generate speech with the "Manbo Storytelling" (Màn Bō Jiǎng Gù Shi) voice style, a popular internet phenomenon widely used in anonymous social media storytelling videos.
This voice style is known for its distinctive, slightly dramatic, and engaging vocal characteristics. The adapter captures the expressive and narrative-driven qualities of this trend, making it suitable for creating storytelling content, narrative videos, and social media short-form content.
中文:
这是一个基于 VoxCPM2 基座模型微调的 LoRA 适配器。它旨在生成具有 "曼波讲故事" 音色风格的语音,该音色是互联网上广泛流行的现象,常见于抖音等社交媒体账号的讲故事类视频中。
这种音色风格以其独特、略带戏剧性且富有感染力的嗓音特征而闻名。本适配器捕捉了这一潮流中富有表现力和叙事感的特点,非常适用于创作讲故事内容、叙事视频以及社交媒体短视频。
📊 训练信息 (Training Info)
- Training Steps: 1000 steps
⚠️ 重要声明 (Important Disclaimer)
English:
- For Educational & Research Purposes Only: This model is released for non-commercial research and learning only.
- No Illegal Use: You are strictly prohibited from using this model for any illegal, harmful, fraudulent, or malicious activities, including but not limited to generating deepfake content, impersonation, or spreading misinformation.
- User Responsibility: Users are solely responsible for complying with all applicable laws and regulations when using this model.
中文:
- 仅供学习与研究用途: 本模型仅供非商业性的研究和学习使用。
- 严禁非法用途: 严格禁止将本模型用于任何非法、有害、欺诈或恶意活动,包括但不限于生成深度伪造内容、冒充他人身份或传播虚假信息。
- 用户责任: 用户在使用本模型时,需自行承担遵守相关法律法规的责任。
Model tree for rav009/VoxCPM2-lora-manbo
Base model
openbmb/VoxCPM2