You can deploy this model and start earning money today!

happyme531/
F5-TTS-RKNN2

Views

No views yet

F5-TTS-RKNN2

在RK3588上运行超高质量的F5-TTS文字转语音/零样本音色克隆模型!

pip install "numpy<2" rknn-toolkit-lite2 jieba torch onnxruntime soundfile pydub pypinyin tqdm

python F5-TTS-ONNX-Inference-rknn2.py

你可以修改F5-TTS-ONNX-Inference-rknn2.py中的文本等参数来生成不同的音频。

pip install "numpy<2" rknn-toolkit2==2.3.0 onnx onnxruntime

bash

python convert_opset.py
python convert_F5_Transformer_opset19.py

由于RKNN2不支持动态输入，这里把序列长度固定为了1536，并通过缩放音频速度来补偿。在差距不大的情况下效果可以接受。
模型中DiT中的RoPE位置编码部分有一个Transpose操作无法在NPU上运行，造成推理速度下降~15%。这个问题应该可以通过修改原模型来解决，但我懒得改了，因为改完之后推理还是会非常慢，因为序列长度实在太长了。
只有DiT部分使用了NPU，其他部分都是CPU推理，但其他部分运行速度快，总体上不会对推理速度有太大影响。

Run the ultra-high-quality F5-TTS text-to-speech / zero-shot voice cloning model on RK3588!

Inference Speed (on RK3588, generating 9 seconds of audio): 11s per iteration, 32 iterations, total time ~352s
Memory Usage (on RK3588): 2.2GB

Clone or download this repository locally. The models are large, ensure you have sufficient disk space.
Install dependencies:

pip install "numpy<2" rknn-toolkit-lite2 jieba torch onnxruntime soundfile pydub pypinyin tqdm
Run:

python F5-TTS-ONNX-Inference-rknn2.py

You can modify parameters such as the text within F5-TTS-ONNX-Inference-rknn2.py to generate different audio.

Download the ONNX model files.
Install dependencies:

pip install "numpy<2" rknn-toolkit2==2.3.0 onnx onnxruntime
Convert the models:

bash
1python convert_opset.py 2python convert_F5_Transformer_opset19.py

Due to RKNN2 limitations with dynamic inputs, the sequence length is fixed at 1536. Audio speed scaling is used to compensate. The effect is acceptable when the difference isn't significant.
A Transpose operation within the RoPE (Rotary Positional Embedding) part of the DiT (Diffusion Transformer) model cannot run on the NPU, causing an approximate 15% decrease in inference speed. This could potentially be resolved by modifying the original model, but I chose not to, as inference would still be very slow due to the extremely long sequence length.
Only the DiT part utilizes the NPU; other parts run on the CPU. However, these CPU parts are fast and do not significantly impact the overall inference speed.