search

k2-fsa / sherpa-onnx

Speech-to-text, text-to-speech, speaker diarization, and VAD using next-gen Kaldi with onnxruntime without Internet connection. Support embedded systems, Android, iOS, Raspberry Pi, RISC-V, x86_64 servers, websocket server/client, C/C++, Python, Kotlin, C#, Go, NodeJS, Java, Swift, Dart, JavaScript, Flutter, Object Pascal, Lazarus, Rust
https://k2-fsa.github.io/sherpa/onnx/index.html
Apache License 2.0
3.7k stars 430 forks source link

encoder_dims does not exist in the metadata #1336

Closed kislay99 closed 2 months ago

kislay99 commented 2 months ago

I'm using this model: sherpa-onnx-zipformer-large-en-2023-06-26.tar.bz2

when building and testing on iPhone I see the following error:

/Users/kislaysingh/offile_tts/sherpa-onnx/sherpa-onnx/csrc/online-zipformer-transducer-model.cc:InitEncoder:98 encoder_dims does not exist in the metadata

I tried making model type both zipformer and zipformer2 getting the same error. Although I'm able to run the following successfully on the iPhone: sherpa-onnx-zipformer-en-2023-04-01

Basically this hugging face playground model does exactly what I'm looking for, I'm guessing it's the large one.

Do I need to use some other configs?

csukuangfj commented 2 months ago

An offline/non-streaming model cannot be used with an OnlineRecognizer.

The error log is expected.

Your usage is not correct.