sherpa_onnx_android 1.12.30
sherpa_onnx_android: ^1.12.30 copied to clipboard
Speech recognition, speech synthesis, and speaker recognition using next-gen Kaldi with onnxruntime without Internet connection.
sherpa-onnx app example #
Flutter examples #
| Functions | URL | Supported Platforms |
|---|---|---|
| Streaming speech recognition | Address | Android, iOS, Linux, macOS, Windows |
| Speech synthesis | Address | Android, iOS, Linux, macOS, Windows |
Pure dart-examples #
Hint: All of the following functions can be used in Flutter, even if some of them are only provided in pure dart api examples.
| Functions | URL | Supported Platforms |
|---|---|---|
| Speaker diarization | Address | macOS, Windows, Linux |
| Streaming speech recognition | Address | macOS, Windows, Linux |
| Non-Streaming speech recognition | Address | macOS, Windows, Linux |
| Text to speech | Address | macOS, Windows, Linux |
| Voice activity detection (VAD) | Address | macOS, Windows, Linux |
| Voice activity detection (VAD) with non-streaming speech recognition | Address | macOS, Windows, Linux |
| Speaker identification and verification | Address | macOS, Windows, Linux |
| Audio tagging | Address | macOS, Windows, Linux |
| Keyword spotter | Address | macOS, Windows, Linux |
| Add punctuations | Address | macOS, Windows, Linux |
| Speech enhancement/denoising | Address GTCRN and DPDFNet (baseline, dpdfnet2, dpdfnet4, dpdfnet8 for 16 kHz ASR, dpdfnet2_48khz_hr for 48 kHz output) |
macOS, Windows, Linux |