TTS_TFLite
TTS_TFLite copied to clipboard
This repository is a collection of TTS Models in TFLite
hifigan lite model on TensorFlowLiteSwift error: input_channels % filter_input_channels != 0 (12!=0)
I'm working with fastspeech2_quant.tflite and hifigan_float16.tflite on TensorFlowLiteSwift The fastspeech output tensor's shape is [1,92,80], and there's an error when resizing input tensor of hifi-gan model with shape [1, 92,...
Has the conversion step from GlowTTS to ONNX been solved? [According to a post on the Mozilla TTS repo](https://github.com/mozilla/TTS/issues/608#issuecomment-781411348_), it appears to be the case. If so, could the steps...
I found in your report that Hifigan Float16 take 0.0036s latency, is it time for inference a sentence or RTF.  I have tried to benchmark with your checkpoint (Tflite...
I tried to follow forward tacotron notebook to convert it into onnx, but cannot get pretrained weights and throws error of "Not found". Also there is no forward tacotron onnx...