Model reference · open weights
fish-speech-1.4 is an open-weight audio or speech model from fishaudio. fish-speech-1.4 (BF16) weighs 989 MB; the smallest configuration that runs it is RTX 3060 12 GB.
Summary of the fishaudio/fish-speech-1.4 model card, 2026-10-01
What it is
| Released by | fishaudio |
|---|---|
| Released | 2024-09-10 |
| VRAM | 989 MB for the weights |
What it runs on
| Card | One stream | Memory |
|---|---|---|
| RTX 3060 12 GB | fits | 11.6 GB |
| RTX 4060 Ti 16 GB | fits | 15.4 GB |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
Fish Speech V1.4 is a leading text-to-speech (TTS) model trained on 700k hours of audio data in multiple languages.
Supported languages:
Please refer to Fish Speech Github for more info. Demo available at Fish Audio.
If you found this repository useful, please consider citing this work:
@article{fish-speech-v1.4,
author = {Shijia Liao, Tianyu Li and others},
title = {Fish-Speech: Leveraging Large Language Models for Advanced Multilingual Text-to-Speech Synthesis},
year = {2024},
journal = {arXiv preprint arXiv:2411.01156},
eprint = {2411.01156},
archivePrefix = {arXiv},
primaryClass = {cs.SD},
url = {https://arxiv.org/abs/2411.01156}
}
This model is permissively licensed under the BY-CC-NC-SA-4.0 license. The source code is released under BSD-3-Clause license.
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
How it works