Release v1.9.24

What's Changed

• Add CTC HLG decoding for JNI by @csukuangfj in #810
• Add function 'tolowerUnicode' in sherpa-onnx-microphone (fix #791) by @daniel-dona in #812
• Add Java API for text-to-speech by @csukuangfj in #811
• Adding temperature scaling on Joiner logits: by @KarelVesely84 in #789
• Fix building wheels for macOS by @csukuangfj in #814
• Fix C# to support Chinese tts models using jieba by @csukuangfj in #815
• Fix a bug for offline paraformer by @csukuangfj in #816
• Add Java API for spoken language identification with whisper multilingual models by @csukuangfj in #817
• Add Java and Kotlin API for punctuation models by @csukuangfj in #818
• Add Java API for audio tagging by @csukuangfj in #820
• Add Java API for speaker identification by @csukuangfj in #822
• Fix typos in JNI TTS by @csukuangfj in #824
• Begin to add node-addon-api for sherpa-onnx by @csukuangfj in #826
• Publish node-addon-api wrapper for sherpa-onnx as npm packages by @csukuangfj in #829
• Update 3dspeaker/export-onnx.py by @chiiyeh in #836
• Upload two more 3d-speaker models by @csukuangfj in #837
• Publish npm package with node-addon-api for Windows by @csukuangfj in #838
• Add links to pre-built APKs and pre-trained models to README. by @csukuangfj in #840
• Publish node-addon-api npm package for linux arm64 by @csukuangfj in #841
• Export NeMo FastConformer Hybrid Transducer-CTC Large Streaming to ONNX. by @csukuangfj in #843
• Export NeMo FastConformer Hybrid Transducer Large Streaming to ONNX by @csukuangfj in #844
• Export non-streaming NeMo faster conformer hybrid transducer and ctc to sherpa-onnx by @csukuangfj in #847
• Add C++ support for non-streaming NeMo fast conformer hybrid transducer ctc (the ctc branch) by @csukuangfj in #848
• Add C++ runtime for non-streaming faster conformer transducer from NeMo. by @csukuangfj in #854
• Solve the issue of missing the last sentence with punctuation by @yh646492956 in #856
• Add C++ support for streaming NeMo CTC models. by @csukuangfj in #857
• Add more streaming ASR methods for node-addon-api by @csukuangfj in #860
• Fix Python TTS examples for models using jieba. by @csukuangfj in #861
• Add Speaker ID demo for C# by @csukuangfj in #862

New Contributors

• @daniel-dona made their first contribution in #812
• @yh646492956 made their first contribution in #856

Full Changelog: v1.9.23...v1.9.24

Release v1.9.23

What's Changed

• fix a typo in building language ID apk by @csukuangfj in #795
• Add jieba for Chinese TTS models by @csukuangfj in #797
• Increase CED's max frame length to 3000 by @csukuangfj in #798
• Fix the last character not being recognized for streaming paraformer … by @csukuangfj in #799
• Refactor TTS Android code to support jieba for Chinese TTS models by @csukuangfj in #800
• wget 续传 by @bubao in #801
• Refactor the JNI interface to make it more modular and maintainable by @csukuangfj in #802
• Fix CI tests by @csukuangfj in #804
• Refactor Java APIs by @csukuangfj in #806
• Add Java API for non-streaming ASR by @csukuangfj in #807
• Add dict_dir arg to c api to support Chinese TTS models using jieba by @csukuangfj in #809

Full Changelog: v1.9.22...v1.9.23

v1.9.22

What's Changed

• Replace torchaudio with soundfile in python-api-examples by @gtf35 in #765
• Add C API for punctuation by @csukuangfj in #768
• Add Kotlin API for audio tagging by @csukuangfj in #770
• Adding warm up for Zipformer2 by @manickavela29 in #766
• Fix display for sherpa-onnx-microphone by @csukuangfj in #773
• Fix code style issues by @csukuangfj in #774
• Add score function to speaker identification by @chiiyeh in #775
• Add Android demo for audio tagging by @csukuangfj in #776
• Add WearOS demo for audio tagging by @csukuangfj in #777
• Add JNI support for spoken language identification by @csukuangfj in #782
• Add Android demo for spoken language identification using Whisper multilingual models by @csukuangfj in #783
• Support CED models by @csukuangfj in #792
• Add Python API example for CED audio tagging. by @csukuangfj in #793
• Release v1.9.22 by @csukuangfj in #794

New Contributors

• @gtf35 made their first contribution in #765

Full Changelog: v1.9.19...v1.9.22

v1.9.19

What's Changed

• Fix a punctuation bug by @csukuangfj in #764

Full Changelog: v1.9.18...v1.9.19

v1.9.18

What's Changed

• Fix building OpenFst on Windows. by @csukuangfj in #744
• Fix go API examples with portaudio on Windows. by @csukuangfj in #746
• Support audio tagging using zipformer by @csukuangfj in #747
• Add C++ microphone examples for audio tagging by @csukuangfj in #749
• Add SHERPA_ONNX_GITHUB by @bubao in #750
• Fix a bug in mean calculation of 'ys_probs' by @aask1357 in #748
• Add Python API and Python examples for audio tagging by @csukuangfj in #753
• Add C API for audio tagging by @csukuangfj in #754
• [feature] Configurable padding length by @manickavela29 in #755
• Use batch size 1 in generating subtitles. by @csukuangfj in #756
• Fix WebAssembly for kws by @csukuangfj in #758
• Support adding punctuations to the speech recogntion result by @csukuangfj in #761
• Add Python API for punctuation models. by @csukuangfj in #762
• Release v1.9.18 by @csukuangfj in #763

New Contributors

• @bubao made their first contribution in #750
• @aask1357 made their first contribution in #748
• @manickavela29 made their first contribution in #755

Full Changelog: v1.9.17...v1.9.18

punctuation-models

Use batch size 1 in generating subtitles. (#756)

audio-tagging-models

Please see also

https://k2-fsa.github.io/sherpa/onnx/audio-tagging/

v1.9.17

What's Changed

• Support heteronyms in Chinese TTS by @csukuangfj in #738
• Add VAD examples using ALSA for recording by @csukuangfj in #739
• Fix releasing GIL by @csukuangfj in #741
• Support Chinese heteronyms on Android for TTS. by @csukuangfj in #742

Full Changelog: v1.9.16...v1.9.17

v1.9.16

What's Changed

• Fix building wasm in CI by @csukuangfj in #720
• Add more piper models for text-to-speech by @csukuangfj in #725
• Fix microphone privacy config by @yujinqiu in #727
• Add language identification swiftui demo by @yujinqiu in #729
• Add HLG decoding for streaming CTC models by @csukuangfj in #731
• Add C API for streaming HLG decoding by @csukuangfj in #734
• return timestamps for WebAssembly by @csukuangfj in #737

Full Changelog: v1.9.15...v1.9.16

v1.9.15

What's Changed

• Add C API for spoken language identification. by @csukuangfj in #695
• Add Swift API for spoken language identification. by @csukuangfj in #696
• Add C# API for spoken language identification by @csukuangfj in #697
• Use a single thread when loading models by @csukuangfj in #703
• Build Android TTS APKs for coqui-ai/TTS models by @csukuangfj in #704
• Support including TTS conditionally. by @csukuangfj in #699
• add openfst.cmake file by @hantengc in #707
• Add Golang API for VAD by @csukuangfj in #708
• Sign released TTS APKs by @csukuangfj in #710
• Add Golang API for spoken language identification. by @csukuangfj in #709
• Fix ios-swift to remove invalid references by @csukuangfj in #713
• delete incorrect logs by @longshiming in #714
• Added progress for callback of tts generator by @leohuang2013 in #712
• Add C API for speaker embedding extractor. by @csukuangfj in #711
• Add Go API for speaker identification by @csukuangfj in #718
• Release v1.9.15 by @csukuangfj in #719

New Contributors

• @hantengc made their first contribution in #707
• @leohuang2013 made their first contribution in #712

Full Changelog: v1.9.14...v1.9.15