Initial

README.md

@@ -0,0 +1,13 @@
+---
+license: mit
+language: en
+tags:
+- text-to-speech
+---
+# Exported [FlowtronTTS](https://arxiv.org/abs/2005.05957) with [WaveGlow](https://arxiv.org/abs/1811.00002) vocoder
+
+This model is intended to be used with [npc-engine](https://github.com/npc-engine/npc-engine).
+
+Fork used for exporting https://github.com/npc-engine/flowtron
+
+Original code https://github.com/NVIDIA/flowtron

backward_flow.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc81544b3d35c63aa8d56b735831bd64467b4a4313f7201ec2190356c3bc4fcb
+size 110745839

config.yml

@@ -0,0 +1,6 @@
+model_type: "FlowtronTTS"
+max_frames: 400
+gate_threshold: 0.5
+sigma: 0.8
+smoothing_window: 3
+smoothing_weight: 0.5

encoder.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fee45c48e4363913bfb97721f3fca7a61e1ce12adfbcba17e29227a83777642
+size 23623159

forward_flow.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1d33e9e6154e28abe816965a61dab61b7c9eefcaad735eb40e827451149f83
+size 110738957

vocoder.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d6710014be0d4dd1196c6f228657c26b6bf84c0ea210764ee388387b7cd67d9
+size 52807590