Release v1.9.26

What's Changed

• Encode hotwords in C++ side by @pkufool in #828
• Fix Go tests by @csukuangfj in #897
• Fix CI tests. by @csukuangfj in #898
• Add Flutter example for speaker identification by @csukuangfj in #894
• Add recording permission for iOS App. by @csukuangfj in #900
• Fix CI for JavaScript and Python APIs. by @csukuangfj in #901
• Fix reading wave files generated by NAudio. by @csukuangfj in #903
• Add Dart API for VAD by @csukuangfj in #904
• Fix CI tests. by @csukuangfj in #907
• fix detecting node-addon packages by @csukuangfj in #908
• Support reading waves from NAudio. by @csukuangfj in #914
• Support Windows arm64 by @csukuangfj in #911
• fix building errors introduced by simple-sentencepiece by @csukuangfj in #915
• Update offline-ctc-greedy-search-decoder.cc by @Dadoou in #917
• Add Flutter GUI example for VAD with a microphone. by @csukuangfj in #905
• 提供设置关键词的api，方便动态调整关键词来进行识别 by @hantengc in #923
• add a new tts piper model by @csukuangfj in #927
• Support not using external buffers for node-addon by @csukuangfj in #925
• Add VAD demo for Java API by @csukuangfj in #928
• Add KWS examples for Java API by @csukuangfj in #930
• Reset encoder states on endpointing for streaming transducer. by @csukuangfj in #924
• fix node-addaon-api for vad by @csukuangfj in #932
• update c-api.h by @RuleNumber1 in #937
• Added tokens, tokens_arr and json for offline recognizer result by @leohuang2013 in #936
• fix: Typo 'maxNumSenetences' in SherpaOnnx.swift by @BrutalCoding in #939
• Split online.cs and offline.csFile by @dfengpo in #941
• Add Dart API for streaming ASR by @csukuangfj in #933
• Add C++ runtime for streaming faster conformer transducer from NeMo. by @sangeet2020 in #889
• Fix nemo streaming transducer greedy search by @csukuangfj in #944
• Wrap punctuation APIs to C#. by @csukuangfj in #945
• Wrap VAD APIs to C# by @csukuangfj in #946
• release v1.9.26 by @csukuangfj in #947
• Fix building for Android by @csukuangfj in #949
• Support customize scores for hotwords by @pkufool in #926
• Add address sanitizer and undefined behavior sanitizer by @csukuangfj in #951

New Contributors

• @Dadoou made their first contribution in #917
• @RuleNumber1 made their first contribution in #937
• @BrutalCoding made their first contribution in #939
• @dfengpo made their first contribution in #941
• @sangeet2020 made their first contribution in #889

Full Changelog: v1.9.25...v1.9.26

Release v1.9.25

What's Changed

• Add node-addon-api for VAD by @csukuangfj in #864
• Fix node addon tests by @csukuangfj in #865
• Add Android APKs for NeMo CTC models. by @csukuangfj in #866
• Add streaming CTC ASR APIs for node-addon-api by @csukuangfj in #867
• Add non-streaming ASR APIs for node-addon-api by @csukuangfj in #868
• Compiler Error and Minor Bug fix by @manickavela29 in #870
• Add TTS for node-addon-api by @csukuangfj in #871
• Add spoken language identification for node-addon-api by @csukuangfj in #872
• Refactor node-addon-api to remove duplicate. by @csukuangfj in #873
• Add speaker identification APIs for node-addon-api by @csukuangfj in #874
• Add audio tagging APIs for node-addon-api by @csukuangfj in #875
• Support adding puncutations to text for node-addon-api by @csukuangfj in #876
• Add keyword spotting API for node-addon-api by @csukuangfj in #877
• Fix sherpa-onnx-node-version in node examples by @csukuangfj in #879
• Update CMakeLists.txt by @linziguan in #881
• Fix Java API examples by @csukuangfj in #883
• Fix a typo in jni by @csukuangfj in #885
• Add tail_paddings to Whisper C API. by @csukuangfj in #886

New Contributors

• @linziguan made their first contribution in #881

Full Changelog: v1.9.24...v1.9.25

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