ModelPrésentéExterne

Neutts Air He

A Hebrew LoRA finetune of neuphonic/neutts-air, adding natural Hebrew text-to-speech with instant voice cloning.

Sourcehuggingface.co/flowty1/neutts-air-he/tree/660eebdff8b58fc109e0386259f0c9d6aa0ef1e0

TypeLoRAModèle de baseneuphonic/neutts-airVersionadapter_model

Source du modèle

Type: LoRA
Taille du fichier: 67,2 MB
Modèle de base: neuphonic/neutts-air
Version: vadapter_model
Mis à jour: 22 sept. 2026
Téléchargements: 12
Tâches: Audio
Source: Hugging Face

Description de la source

NeuTTS Air — Hebrew (LoRA)

A Hebrew LoRA finetune of neuphonic/neutts-air, adding natural Hebrew text-to-speech with instant voice cloning.

Base model: neuphonic/neutts-air (Apache-2.0)
Adapter type: LoRA (PEFT), rank 32, alpha 64, rsLoRA
Language: Hebrew (he)
Inference example / demo: flowtyone/neutts-air-he-lora on GitHub

This adapter must be applied on top of the base backbone; it does not contain the full model weights. For Hebrew text it is paired with the thewh1teagle/phonikud grapheme-to-phoneme (diacritization) model and neuphonic/neucodec.

Usage

import soundfile as sf
from peft import PeftModel
from transformers import AutoModelForCausalLM

# Minimal sketch — see the GitHub inference example for the full pipeline
# (phonemization with phonikud + audio codec with neucodec).
backbone = AutoModelForCausalLM.from_pretrained("neuphonic/neutts-air")
model = PeftModel.from_pretrained(backbone, "flowty1/neutts-air-he")

The full inference pipeline (reference encoding, phonemization, and waveform decoding) is provided in the GitHub repository, which loads this adapter directly by repo id:

from lib.neutts import NeuTTSAir

tts = NeuTTSAir(
    backbone_path="neuphonic/neutts-air",
    lora_path="flowty1/neutts-air-he",
    codec_path="neuphonic/neucodec",
    codec_encoder_path="facebook/w2v-bert-2.0",
    g2p_model_path="thewh1teagle/phonikud",
    device="auto",
)

Reference audio guidelines

For best voice-cloning quality, use a clean, mono reference clip of 3–15 seconds of natural, continuous speech, together with an accurate transcription.

Acknowledgements

Neuphonic — NeuTTS Air and NeuCodec.
Phonikud / DICTA — Hebrew diacritization.

License

This LoRA adapter is released under the Apache-2.0 license, matching the base model neuphonic/neutts-air.

Sources

1 source

Hugging FacePrincipalVersion 660eebdff8b58fc109e0386259f0c9d6aa0ef1e0

huggingface.co/flowty1/neutts-air-he/tree/660eebdff8b58fc109e0386259f0c9d6aa0ef1e0

Vérifié 22 sept.

Artefacts du modèle

1 artefact

adapter_model.safetensors

safetensors · 67,2 MB · SHA-256 0adc70266928…fc55 · Hugging Face

Télécharger

Extraits de sources

2 extraits

Readme22 sept.huggingface.co/flowty1/neutts-air-he

--- license: apache-2.0 base_model: neuphonic/neutts-air library_name: peft language: - he pipeline_tag: text-to-speech tags: - text-to-speech - tts - hebrew - neutts - neutts-air - lora - peft - voice-cloning --- # NeuTTS Air — Hebrew (LoRA) A Hebrew LoRA finetune of [`neuphonic/neutts-air`](https://huggingface.co/neuphonic/neutts-air), adding natural Hebrew text-to-speech with instant voice cloning. - **Base model:** [`neuphonic/neutts-air`](https://huggingface.co/neuphonic/neutts-air) (Apache-2.0) - **Adapter type:** LoRA (PEFT), rank 32, alpha 64, rsLoRA - **Language:** Hebrew (`he`) - **Inference example / demo:** [`flowtyone/neutts-air-he-lora`](https://github.com/flowtyone/neutts-air-he-lora) on GitHub This adapter must be applied on top of the base backbone; it does not contain the full model weights. For Hebrew text it is paired with the [`thewh1teagle/phonikud`](https://huggingface.co/thewh1teagle/phonikud) grapheme-to-phoneme (diacritization) model and [`neuphonic/neucodec`](https://huggingface.co/neuphonic/neucodec). ## Usage ```python import soundfile as sf from peft import PeftModel from transformers import AutoModelForCausalLM # Minimal sketch — see the GitHub inference example for the full pipeline # (phonemization with phonikud + audio codec with neucodec). backbone = AutoModelForCausalLM.from_pretrained("neuphonic/neutts-air") model = PeftModel.from_pretrained(backbone, "flowty1/neutts-air-he") ``` The full inference pipeline (reference encoding, phonemization, and waveform decoding) is provided in the [GitHub repository](https://github.com/flowtyone/neutts-air-he-lora), which loads this adapter directly by repo id: ```python from lib.neutts import NeuTTSAir tts = NeuTTSAir( backbone_path="neuphonic/neutts-air", lora_path="flowty1/neutts-air-he", codec_path="neuphonic/neucodec", codec_encoder_path="facebook/w2v-bert-2.0", g2p_model_path="thewh1teagle/phonikud", device="auto", ) ``` ## Reference audio guidelines For best voice-cloning quality, use a clean, mono reference clip of 3–15 seconds of natural, continuous speech, together with an accurate transcription. ## Acknowledgements - [Neuphonic](https://huggingface.co/neuphonic) — NeuTTS Air and NeuCodec. - [Phonikud](https://huggingface.co/thewh1teagle/phonikud) / [DICTA](https://huggingface.co/dicta-il) — Hebrew diacritization. ## License This LoRA adapter is relea...

Note22 sept.huggingface.co/flowty1/neutts-air-he

Source context: 12 downloads · 1 likes · Pipeline text-to-speech · Library peft · Repo flowty1/neutts-air-he