v1.9.12

What's Changed

• Add Python API for keyword spotting by @pkufool in #576
• Fix building wheels by @csukuangfj in #620
• Add WebAssembly demo for streaming trilingual Paraformer (Chinese+Cantonese+English) by @csukuangfj in #618
• Support Android NNAPI. by @csukuangfj in #622
• fixed variable's spell num_trailing_blanks by @lonngxiang in #623
• Support VITS models from icefall. by @csukuangfj in #625
• support onnxruntime v1.17.1 by @csukuangfj in #624
• Refactor WebAssembly for nodejs by @csukuangfj in #626
• Add WebAssembly for NodeJS. by @csukuangfj in #628
• Fix building wheels from source. by @csukuangfj in #632
• Support Ubuntu 18.04 and CentOS 7 by @csukuangfj in #635
• Allow to not use pre-installed onnxruntime libs. by @csukuangfj in #636
• Support using T-head-Semi/csi-nn2 for RISC-V by @csukuangfj in #637
• Fix deletions at the tail when using hotwords in streaming model by @pkufool in #638
• Add Python ASR examples with alsa by @csukuangfj in #646
• Add microphone streaming ASR example for C API by @csukuangfj in #650
• Expose 'language' 'task' 'tailPaddings' in OfflineWhisperModelConfig by @GaryLaurenceauAva in #643
• Support user provided data in tts callback. by @csukuangfj in #653
• Use onnxruntime 1.17.1 for iOS. by @csukuangfj in #654
• c++ api for keyword spotter by @xinhecuican in #642
• Print the time about the first message in tts. by @csukuangfj in #655
• Release v1.9.12 by @csukuangfj in #661

New Contributors

• @lonngxiang made their first contribution in #623
• @GaryLaurenceauAva made their first contribution in #643
• @xinhecuican made their first contribution in #642

Full Changelog: v1.9.11...v1.9.12