Add Java and Kotlin API for NeMo Canary models (#2359)

Add support for the NeMo Canary model in both Java and Kotlin APIs, wiring it through
JNI and updating examples and CI.

- Introduce OfflineCanaryModelConfig in Kotlin and Java with builder patterns
- Extend OfflineRecognizer to accept and apply the new canary config via setConfig
- Update JNI binding (GetOfflineConfig) and getOfflineModelConfig mapping (type 32), 
   plus examples and CI workflows
This commit is contained in:
Fangjun Kuang
2025-07-08 13:45:26 +08:00
committed by GitHub
parent df4615ca1d
commit 103e93d9f6
12 changed files with 363 additions and 11 deletions

View File

@@ -24,11 +24,18 @@ This directory contains examples for the JAVA API of sherpa-onnx.
```bash
./run-non-streaming-decode-file-dolphin-ctc.sh
./run-non-streaming-decode-file-fire-red-asr.sh
./run-non-streaming-decode-file-moonshine.sh
./run-non-streaming-decode-file-nemo-canary.sh
./run-non-streaming-decode-file-nemo.sh
./run-non-streaming-decode-file-paraformer.sh
./run-non-streaming-decode-file-sense-voice.sh
./run-non-streaming-decode-file-tele-speech-ctc.sh
./run-non-streaming-decode-file-transducer-hotwords.sh
./run-non-streaming-decode-file-transducer.sh
./run-non-streaming-decode-file-whisper-multiple.sh
./run-non-streaming-decode-file-whisper.sh
./run-non-streaming-decode-file-nemo.sh
./run-non-streaming-decode-file-zipformer-ctc.sh
```
## Non-Streaming Speech recognition with homophone replacer