zipformer-asr-rknn – AI Model by voicialex | AlphaNeural AI

You can deploy this model and start earning money today!

voicialex/
zipformer-asr-rknn

like

0

Views

No views yet

zipformer-asr-rknn

Streaming ASR (Automatic Speech Recognition) — bilingual Chinese/English, based on sherpa-onnx Zipformer.

Models

ONNX (CPU, x86_64 / aarch64)

File	Description
`onnx/encoder-epoch-99-avg-1.onnx`	Streaming encoder
`onnx/decoder-epoch-99-avg-1.onnx`	Decoder
`onnx/joiner-epoch-99-avg-1.onnx`	Joiner
`onnx/tokens.txt`	Token vocabulary

RKNN (RK3588 NPU)

File	Description
`rknn/encoder.rknn`	Streaming encoder
`rknn/decoder.rknn`	Decoder
`rknn/joiner.rknn`	Joiner
`rknn/tokens.txt`	Token vocabulary

Source

Converted from sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20.

Runtime

Used by the buddy ASR pipeline via sherpa-onnx or RKNN SDK.