This release fine-tunes the text-to-token and token-to-mel stages of
CosyVoice2-0.5B
(FunAudioLLM / Alibaba, Apache-2.0) for Amharic. It keeps the base model's
reference-audio voice cloning, which uses about ten seconds of audio. The
model produces intelligible Amharic; frequent hesitation sounds are its main
quality limitation.
The tokenizer represents Ge'ez text without unknown tokens, but the stock
checkpoint did not produce intelligible Amharic in our tests. The vocoder,
speech tokenizer and speaker encoder are unchanged.
amharic_text.py is the text normalizer used for training
labels and at inference.
Hear it
Each pair uses the same sentence, reference audio, settings and seed.
#
Test case
Stock CosyVoice2
+ Gabar fine-tune
1
Ordinary prose
2
Prose, ፥ punctuation
3
Prose (ejectives ቡ/ጅ)
4
Numbers + ዓ.ም. date abbreviation
5
ዶ/ር title abbreviation
6
Question intonation
7
Technical prose
8
Mixed punctuation + question
The players render on Hugging Face only. On GitHub the clips are in
demo/ — stock:
1 · 2 · 3 ·
4 · 5 · 6 ·
7 · 8; fine-tuned:
1 · 2 · 3 ·
4 · 5 · 6 ·
7 · 8.
The weights (llm.pt, flow.pt, ~2 GB) are on
Hugging Face;
everything else is mirrored on GitHub.
Texts: demo/sentences.txt. Reference voice:
demo/reference.wav, a consenting member of the Gabar team
(). Both models got the text after
amharic_text.normalize, cross-lingual mode (reference audio only), seed 1234. No outputs were
selected from multiple attempts. Known weaknesses are under Limitations.
Files in this release
Two files: llm.pt (the text → speech-token language model) and flow.pt
(speech tokens → mel), both full weights, both fine-tuned from the stock
CosyVoice2-0.5B checkpoint. You combine them with the stock model, which the
loader downloads from FunAudioLLM/CosyVoice2-0.5B at the pinned revision;
the HiFT vocoder, the speech tokenizer, the campplus speaker encoder and the
Qwen text tokenizer are stock and unmodified.
This is not a standalone model and not an adapter: the two stages are
complete replacement weights. This checkpoint is intended only for Amharic;
use the original base checkpoint for its supported languages.
Amharic text front-end
amharic_text.py builds the training labels and the
loader runs it on every input, so training and inference see the same text.
It converts Ge'ez numerals, digits, percentages and times to words, expands
about 100 abbreviations, collapses homophones, and reduces punctuation to
።፣፤?!.
The training set contains the three sources listed below. Every clip's filename starts with its corpus prefix. We
audited the assembled training directory before training; the report is
included in this repository as is
(audit/corpus_audit.txt). Both stages were
trained from the stock CosyVoice2 checkpoint on exactly that directory.
WaxalNLP comes as 48 kHz and Common Voice as 32/48 kHz MP3; both were
resampled to 24 kHz.
Licence: why CC-BY-SA-4.0
WaxalNLP is licensed under CC BY-SA 4.0. The fine-tuned weights are released
under CC BY-SA 4.0. Third-party components retain their original licences;
the base model is Apache-2.0. Credit to Digital Umuganda, the WaxalNLP
contributors, and the Common Voice contributors.
Architecture
Base: CosyVoice2-0.5B, FunAudioLLM/CosyVoice2-0.5B at revision
eec1ae6c79877dbd9379285cf8789c9e0879293d; code FunAudioLLM/CosyVoice at commit 074ca6dc9e80a2f424f1f74b48bdd7d3fea531cc.
The loader pins these revisions because the fine-tuned stages depend on the
corresponding tokenizer, vocoder, and speaker encoder.
component
treatment
LLM (Qwen-based text → speech-token, 0.5 B)
full fine-tune from the stock llm.pt; shipped
Flow (speech tokens → mel, flow matching)
full fine-tune from the stock flow.pt, non-streaming batches; shipped
HiFT vocoder (hift.pt)
stock, untouched, not shipped
Speech tokenizer (speech_tokenizer_v2.onnx)
stock, frozen (its quantization sets a ceiling on ejectives/gemination — see Limitations)
campplus speaker encoder / Qwen text tokenizer
stock, untouched, not shipped
Evaluation
Held-out set: we evaluated 100 clips sampled from a
104-clip speaker-disjoint held-out partition of the same
corpus, split before training by a seeded rule, so whole speakers are held
out and none of their sentences appear in training. Checked independently of the
trainer's own split: HELD OUT: eval ∩ train = ∅ at clip, speaker and sentence level; all eval stems are cv_/ih_/wxl_.
(audit/holdout_verify.txt). Both models ran on
the same clips with the same per-clip reference audio, in cross-lingual mode
(reference audio only — handing either model the reference transcript would
let it copy the answer), through the same code: the released wrapper with our
weights, and the identical assembly with the stock llm.pt/flow.pt.
metric
stock CosyVoice2
+ Gabar fine-tune
Amharic CER ↓ (Meta omniASR-CTC-3B)
0.939 (over 67/100 clips)
0.094 (over 99/100 clips)
UTMOS ↑ (naturalness MOS predictor)
2.177
2.787
ECAPA cosine ↑ (speaker similarity to reference)
0.397
0.758
generation failures (empty / <0.5 s / error)
0.0%
0.0%
Per-clip numbers: audit/eval/. CER method: Meta's stock
omniASR-CTC-3B transcribes
the generated audio; reference and hypothesis are homophone-collapsed and
punctuation-stripped before comparison. Same ASR and normalization for both
models. UTMOS was trained on English MOS ratings and is a rough signal only.
Means cover clips that produced audio; generation failures are reported
separately.
CER was calculated only where the ASR returned a usable transcript: 67 of 100
stock outputs and 99 of 100 fine-tuned outputs. Missing transcripts are
excluded from the mean, so the coverage counts should be read alongside the
CER values.
In our listening review: the stock outputs were not intelligible Amharic. The fine-tuned outputs were intelligible, but frequent "emm" and "uh" fillers made them sound unnatural.
Usage
This release uses a pinned CosyVoice checkout rather than a PyPI package:
1from huggingface_hub import hf_hub_download
2import importlib.util, torchaudio
34# the loader + text front-end ship in this repo5spec = importlib.util.spec_from_file_location(6"amharic_tts", hf_hub_download("gabar-tech/cosyvoice2-amharic","amharic_tts.py"))7amharic_tts = importlib.util.module_from_spec(spec); spec.loader.exec_module(amharic_tts)89tts = amharic_tts.load_amharic_tts(cosyvoice_repo="CosyVoice")# downloads base (pinned) + fine-tune10wav = tts.generate(11"ሰላም! ይህ ከጽሑፍ በቀጥታ የተፈጠረ የአማርኛ ድምፅ ነው። ዛሬ ነሐሴ 11 ቀን 2018 ዓ.ም. ነው።",12 audio_prompt_path="reference.wav",# ~10 s of the voice to clone, with consent13 prompt_text="<exact transcript of reference.wav>")# optional but much better14torchaudio.save("out.wav", wav, tts.sr)# 24 kHz, PerTh-watermarked by the loader
Or from a checkout:
python amharic_tts.py "ሰላም ዓለም።" --ref reference.wav --ref-text "<its transcript>" --out out.wav.
generate() normalizes the text, splits at sentence-final marks, merges
very short chunks (they are less stable during synthesis), synthesizes each sentence
against the reference and joins them. We ran the snippet above as written
on NVIDIA A40, Ubuntu 22.04.5 LTS, python 3.11.10, torch 2.3.1+cu121, CosyVoice@074ca6dc (its requirements.txt) + resemble-perth, before publishing.
Watermarking
The included loader applies Resemble's PerTh watermark (resemble-perth,
MIT) to every waveform generate() returns. The watermark is added by the
loader rather than embedded in the weights, so it can be bypassed through
another inference path and cannot be guaranteed for every use of the model.
We ran the public detector
over every demo clip from both models and the held-out eval outputs, with
the natural reference recording as a negative control: present on all 116 checked files (detector confidence ≥ 0.5 on every generated file; the natural reference recording scores 0.0, so the detector is discriminating)
(audit/watermark_verify.txt). If you build
on this, keep the watermark in.
Intended use
Amharic speech interfaces, audiobooks, education, accessibility, media
production, with the consent of whoever's voice you clone.
Research on low-resource TTS and Ethiopian language technology.
Out of scope
Cloning someone's voice without their informed consent.
Political persuasion, impersonating public figures, fraud (voice
authentication included), harassment.
Other languages. Use the original base checkpoint for those.
Anything safety-critical or broadcast without a human listening first.
Limitations
Gemination (consonant length; contrastive in Amharic, unwritten in normal
spelling) isn't marked; the model reads minimal pairs from context and
will get some wrong.
The speech tokenizer is stock and frozen. Whatever detail it fails to
capture about ejectives or gemination, no fine-tune of the stages behind
it can recover.
Numbers, abbreviations and Ge'ez numerals are expanded by
amharic_text.py. Skip it and the model gets them raw.
Zero-shot mode (reference transcript given) is noticeably more stable than
cross-lingual mode (no transcript). If you have the transcript, pass it.
Very short inputs can come out unstable; the loader merges short chunks,
but single words are still best put in a sentence.
The model inserts an "emm" or "uh" before some words.
This is its main quality limitation. About 96% of the 193.63 hours
of training audio is spontaneous image-description speech rather than read
speech, and that portion runs about 21% slower by characters per second.
This speaking style is a likely contributor; we have not isolated the cause.
Expect conversational rather than neutral read-aloud delivery.
Amharic only.
Risks and misuse
This model can clone a voice from roughly ten seconds of reference audio, so
it can be misused for impersonation or fraud. Use it only with informed
consent and human review.
Every voice in the training data was recorded under a licence that permits
this use.
Attribution
FunAudioLLM / Alibaba, for CosyVoice2 (Apache-2.0), the base model.
Resemble AI, for the public PerTh watermarker (MIT).
Digital Umuganda and the WaxalNLP contributors, Amharic speech data
(CC-BY-SA-4.0).
Mozilla Common Voice contributors, Amharic speech data (CC0-1.0).
Citation
bibtex
1@misc{gabar2026cosyvoice2amharic,
2 title = {CosyVoice2 Amharic: an Amharic fine-tune of CosyVoice2-0.5B},
3 author = {{Gabar Technologies}},
4 year = {2026},
5 url = {https://huggingface.co/gabar-tech/cosyvoice2-amharic}
6}
