Meorav POC

What it is

Give it a line of English with Hebrew words in Latin letters, and it reads the line aloud, with the Hebrew said the Israeli way rather than spelled out as English:

Yalla, tachles: Bituach Leumi wants the form by Thursday.

Meorav is a full fine-tune (every weight updated) of Qwen3-TTS-12Hz-1.7B-Base. Qwen3-TTS reads raw text through its tokenizer, with no phonemiser in between, and a full fine-tune lets the model move off the American accent it starts with. Two earlier attempts at this voice, an Orpheus fine-tune and a zero-shot clone, came out American: a zero-shot clone copies the timbre but takes its prosody from the base model.

Meorav is Hebrew for "mixed", which is what code-switching is, and meorav yerushalmi is the Jerusalem mixed grill.

The other route to the same goal is the deterministic English-Hebrew Code-Switching TTS Pipeline (romanized Hebrew, then Hebrew script with niqqud, then a Hebrew voice for each Hebrew span; see the token classifier); Meorav is one model that learned the Hebrew words from recordings instead.

Status

Meorav PoC-1 POC is the first training run (first-test, 2026-10-22): 657 read takes, 50 minutes after trimming, from 48 kHz / 24-bit recordings resampled to 24 kHz for training. Learning rate 7e-6, 3 epochs, batch 2 with gradient accumulation 4, on one L40S on Modal. It is published, open, as Hebrish/TTS-V1.

No benchmark exists yet, and the listening verdict is pending. PoC-1 is a test of what a small amount of data can do, not a finished voice; expect rough edges.

Next is Meorav v1.0, the first mature release, on about 2.5 to 3 hours of recordings including free speech. The Hugging Face model takes the Meorav name then; Hebrish/TTS-V1 predates the name. Later retrains are v1.1, v1.2 and so on, as with the other Hebrish lines.

Recordings

Meorav is trained on one speaker, Daniel Rosehill, and speaks in his voice. The takes are in Hebrish/Hebrish-TTS-Voice-Recordings (CC BY-NC 4.0): mono WAV at 48 kHz / 24-bit, with no browser processing, read from prompt sets that cover English phonetics, everyday conversation, English sentences carrying Hebrew words and Israeli names and places, numbers, tech vocabulary and free speech.

They are recorded in a dedicated quiet-room voice-clone mode of the phone capture app that also collects Hebrish's speech recognition data. The two need opposite conditions: speech recognition training uses every take, including deliberately noisy ones (an air conditioner, people talking), while a voice clone needs a quiet room. Only voice-clone takes that pass the level checks train Meorav.

Use it

import torch, soundfile as sf
from qwen_tts import Qwen3TTSModel

tts = Qwen3TTSModel.from_pretrained("Hebrish/TTS-V1", device_map="cuda:0", dtype=torch.bfloat16)
wavs, sr = tts.generate_custom_voice(text="Yalla, I'll ring you after the match.", speaker="daniel", language="English")
sf.write("out.wav", wavs[0], sr)

Needs qwen-tts (tested with 0.1.1) and a GPU. Numbers are spelled out in the training text, so write them out in prompts too ("four hundred and twelve"). From the model card.

The training pipeline is public: Hebrish-ASR/Meorav-TTS has the data preparation, the Modal training stages, the run log, and a patched copy of Qwen3-TTS's fine-tuning script with fixes useful to anyone fine-tuning it. Its README explains how to adapt it to another voice.

At a glance

Repository

Hugging Face issues a DOI per repository, so each format can be cited on its own. Repositories, sizes and DOIs checked on the Hub on 2026-10-22.

Versions

Cite

Meorav has no DOI yet; cite the repository, Hebrish/TTS-V1.

@misc{rosehill_meorav_2026,
  author    = {Daniel Rosehill},
  title     = {Meorav PoC-1 (Hebrish/TTS-V1)},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/Hebrish/TTS-V1}
}

Licence and access

Speech synthesis · All Hebrish models