Fishspeech S2 Pro "Miku Edition" V2

A Miku Nakano voice fine-tune of Fish Audio S2 Pro, made as part of my Quintessential Quintuplets fan-model collection.

Give it Japanese text and the included reference to generate new speech in this character's voice. Both the reference and the line affect pace, expression, and similarity.

Built with Fish Audio. This is an unofficial fan project, with no affiliation or endorsement from the creators, performers, or franchise.

Blind listening results: Blind listening tests found V2 to be a dramatic improvement over the base model and V1.

Full merged model

This repository contains the complete BF16 merged model. The trained Miku adaptation is already in model.safetensors. No separate LoRA adapter, base-weight download, or merge step is required. Load it as a full model with lora_config=None; do not apply an adapter again.

File	Purpose
model.safetensors	Full merged text-to-semantic weights, 9.92 GB
codec.pth	Original pinned Fish Speech audio codec, 1.87 GB
config.json and tokenizer files	Native dual_ar configuration and tokenizer
infer.py	Wrapper around the pinned Fish Speech inference CLI
references/*.wav, .txt, .npy	Multiple matching reference recordings, transcripts, and codec tokens
LICENSE.md, NOTICE	License and attribution/modification notice

The training corpus, optimizer states, checkpoint files, and standalone adapters are not included.

References

Default: g1_mi_00002. All reference recordings are preserved unchanged; each includes its matching transcript and precomputed codec tokens.

g1_mi_00002 (default)

7.34 seconds. WAV · Transcript · Tokens

フータローへのお礼をかねてきたんだから、後の段取りは私たちに任せて、のんびりして。

g1_mi_00004

7.89 seconds. WAV · Transcript · Tokens

あ、そうだ。その前に日焼け止め塗らないと。ねえ、風太郎、塗ってくれる？

g2_mi_00014

8.31 seconds. WAV · Transcript · Tokens

風太郎にしては早いかも。正直卒業くらいまで待たされるんじゃないかって思ってた。

Choose another included reference with --reference; keep its transcript and tokens together:

python checkpoints/miku-v2/infer.py --text "今日は何をして過ごしましょうか。" --reference g1_mi_00004 --output miku.wav

Changing the reference can change delivery. The voice is not guaranteed to match the character exactly on every line.

Use the model

Use the compatible Fish Speech fork at commit 166940c301b8b4650b3110397a38bc68727a3afa. This is a native Fish Speech dual_ar model, not a drop-in Transformers AutoModel checkpoint.

Install the fork and a PyTorch build compatible with your hardware using that revision's instructions. The pinned project specifies PyTorch 2.8.0; accelerator support depends on your runtime.

git clone https://github.com/TachibanaKimika/fish-speech.git
cd fish-speech
git checkout 166940c301b8b4650b3110397a38bc68727a3afa
# Follow this revision's platform-specific dependency installation instructions.

Authenticate with hf auth login using an account with access to this private repository, then download the complete model:

hf download Skttttt/Fishspeech-S2-Pro-Miku-Edition-V2 --local-dir checkpoints/miku-v2

From the installed Fish Speech source directory:

python checkpoints/miku-v2/infer.py --text "今日は何をして過ごしましょうか。" --output miku.wav

Defaults: included reference, seed 42, BF16, 4,096-token model context, top-p 0.7, top-k 30, temperature 0.7, and at most 1,024 new audio tokens. The helper saves the WAV and generated codec tokens beside it in an output-stem_codes folder. Resource use depends on the text. Downloading the repository does not load or run inference.

Fine-tuning audio

Fine-tuned on approximately 4.17 hours of real recorded Miku speech, excluding validation and test audio.

Credits and license

Thanks to Fish Audio, Negi Haruba, and the cast and crew of The Quintessential Quintuplets.

The model and derivative weights follow the Fish Audio Research License. See NOTICE for attribution and modification details. Source recordings, performances, character, and franchise remain subject to their respective rights; this project does not claim ownership of them.