default noise_scale=0.0 (stable long-sequence decoding)

add mel sharpen (before soft-gate): restore aspiration detail

strip trailing silence padding in vocoder_chunked (single-chunk path)

Use inline ONNX (single-file), fix tail_stretch dtype

Release 1.0.0: Chinese TTS (Inflect-Micro-v2 arch + BigVGAN), ONNX + AX650 axmodel + checkpoints