Commits · 2024e966399bbf68171ae006f9b940ec43667c16 · xuning / sherpaonnx

13 Jan, 2024 2 commits
- Add C++ runtime for speaker verification models from NeMo (#527) · 2024e966
  2024e966 浏览文件
  
  Fangjun Kuang authored 2024-01-13 21:42:09 +0800
- Export speaker verification models from NeMo to ONNX (#526) · 68a525a0
  68a525a0 浏览文件
  
  Fangjun Kuang authored 2024-01-13 19:49:45 +0800
11 Jan, 2024 2 commits
- Add C++ runtime for models from 3d-speaker (#523) · afc81ec1
  afc81ec1 浏览文件
  
  Fangjun Kuang authored 2024-01-11 19:10:30 +0800
- Fix publishing nuget packages. (#525) · ec728ff7
  ec728ff7 浏览文件
  
  Fangjun Kuang authored 2024-01-11 18:54:23 +0800
10 Jan, 2024 1 commit
- Support exporting models to onnx from 3D-Speaker (#522) · 07e2b9a3
  07e2b9a3 浏览文件
  
  Fangjun Kuang authored 2024-01-10 21:09:45 +0800
09 Jan, 2024 1 commit
- Add runtime support for wespeaker models (#516) · 55266918
  55266918 浏览文件
  
  Fangjun Kuang authored 2024-01-09 22:06:08 +0800
05 Jan, 2024 1 commit
- Use NDK 22.1 for android build (#518) · 902b2189
  902b2189 浏览文件
  
  Fangjun Kuang authored 2024-01-05 20:34:01 +0800
04 Jan, 2024 2 commits
- Use high_freq -400 in computing fbank features. (#515) · 0be71a31 ...
  0be71a31 Browse Directory
```
Fixes #514
```
  Fangjun Kuang authored 2024-01-04 12:39:06 +0800
- Fix #510 (#513) · 547a22f7
  547a22f7 浏览文件
  
  Fangjun Kuang authored 2024-01-04 12:32:19 +0800
03 Jan, 2024 2 commits
- Fix Byte BPE string results for Python. (#512) · e215d0c3 ...
  e215d0c3 浏览文件
```
It ignores invalid UTF8 strings.
```
  Fangjun Kuang authored 2024-01-03 16:03:24 +0800
- Add missing field for two-pass APK. (#511) · d0114217
  d0114217 浏览文件
  
  Fangjun Kuang authored 2024-01-03 12:51:54 +0800
01 Jan, 2024 1 commit
- Build text-to-speech engine APKs (#509) · 581eceb4
  581eceb4 浏览文件
  
  Fangjun Kuang authored 2024-01-01 12:44:20 +0800
31 Dec, 2023 1 commit
- Replace Android system TTS engine (#508) · d7e10bb3
  d7e10bb3 浏览文件
  
  Fangjun Kuang authored 2023-12-31 23:02:35 +0800
22 Dec, 2023 1 commit

Support streaming zipformer CTC (#496) · e475e750 ...

* Support streaming zipformer CTC

* test online zipformer2 CTC

* Update doc of sherpa-onnx.cc

* Add Python APIs for streaming zipformer2 ctc

* Add Python API examples for streaming zipformer2 ctc

* Swift API for streaming zipformer2 CTC

* NodeJS API for streaming zipformer2 CTC

* Kotlin API for streaming zipformer2 CTC

* Golang API for streaming zipformer2 CTC

* C# API for streaming zipformer2 CTC

* Release v1.9.6

authored 2023-12-22 13:46:33 +0800

20 Dec, 2023 2 commits
- Release Python GIL in C++ class constructor (#493) · 7634f5f0
  7634f5f0 Browse Directory
  
  Fangjun Kuang authored 2023-12-20 15:54:32 +0800
- Fix whisper test script for the latest onnxruntime. (#494) · ef8d112a
  ef8d112a 浏览文件
  
  Fangjun Kuang authored 2023-12-20 11:12:12 +0800
15 Dec, 2023 3 commits
- Keep multiple threads from calling into espeak-ng at the same time (#489) · 03ff9db5
  03ff9db5 Browse Directory
  
  Fangjun Kuang authored 2023-12-15 17:44:33 +0800
- Print informative error messages for sherpa-onnx-alsa on errors. (#486) · ad72e7af
  ad72e7af 浏览文件
  
  Fangjun Kuang authored 2023-12-15 11:10:39 +0800
- Fix CI (#485) · 33c03f78
  33c03f78 Browse File
  
  Fangjun Kuang authored 2023-12-15 10:25:03 +0800
14 Dec, 2023 1 commit
- fix building linux x86 wheels (#484) · 9ff6185b
  9ff6185b Browse File
  
  Fangjun Kuang authored 2023-12-14 21:37:40 +0800
13 Dec, 2023 2 commits
- Play generated audio using alsa for TTS (#482) · b18812ce
  b18812ce 浏览文件
  
  Fangjun Kuang authored 2023-12-13 22:28:03 +0800
- Add two GLaDOS TTS models (#481) · 9829d7c4
  9829d7c4 浏览文件
  
  Fangjun Kuang authored 2023-12-13 15:40:07 +0800
10 Dec, 2023 1 commit
- Fix android tts audio buffer size and fix CI. (#478) · 80d01923
  80d01923 浏览文件
  
  Fangjun Kuang authored 2023-12-10 18:25:50 +0800
09 Dec, 2023 1 commit
- Support playing as it is generating for Android (#477) · 0f053d80
  0f053d80 浏览文件
  
  Fangjun Kuang authored 2023-12-09 16:36:38 +0800
08 Dec, 2023 4 commits
- Fix releasing go packages (#476) · cae0231f
  cae0231f 浏览文件
  
  Fangjun Kuang authored 2023-12-09 00:07:52 +0800
- convert wespeaker models to sherpa-onnx (#475) · aef74c51
  aef74c51 Browse Directory
  
  Fangjun Kuang authored 2023-12-08 19:32:29 +0800
- Give an informative log for whisper on exceptions. (#473) · 0e23f826
  0e23f826 浏览文件
  
  Fangjun Kuang authored 2023-12-08 14:33:59 +0800
- Support distil-small.en whisper (#472) · 868c339e
  868c339e Browse Directory
  
  Fangjun Kuang authored 2023-12-08 11:59:20 +0800
07 Dec, 2023 1 commit
- Remove the 30-second constraint from whisper. (#471) · 3ae984f1
  3ae984f1 浏览文件
  
  Fangjun Kuang authored 2023-12-07 17:47:08 +0800
06 Dec, 2023 3 commits
- Release v1.9.0 (#470) · a7d69359
  a7d69359 Browse File
  
  Fangjun Kuang authored 2023-12-06 19:46:50 +0800
- Support Ukrainian VITS models from coqui-ai/TTS (#469) · d3416141
  d3416141 Browse Directory
  
  Fangjun Kuang authored 2023-12-06 19:37:11 +0800
- Use espeak-ng for coqui-ai/TTS VITS English models. (#466) · 23cf92da
  23cf92da 浏览文件
  
  Fangjun Kuang authored 2023-12-06 11:00:38 +0800
04 Dec, 2023 2 commits
- Fix building for .Net (#463) · 3b90e85e
  3b90e85e 浏览文件
  
  Fangjun Kuang authored 2023-12-04 19:27:55 +0800
- Support playing generated audio as it is generating for MFC. (#462) · 73afa024 ...
  73afa024 浏览文件
```
* Support playing generated audio as it is generating for MFC.

* support espeak-ng-data
```
  Fangjun Kuang authored 2023-12-04 14:23:38 +0800
03 Dec, 2023 1 commit
- Break text into sentences for tts. (#460) · 86b4be52 ...
  86b4be52 浏览文件
```
This is for models that are not using piper-phonemize as their front-end.
```
  Fangjun Kuang authored 2023-12-03 11:50:25 +0800
02 Dec, 2023 1 commit
- Play generated audio as it is generating. (#457) · 99ff6a83
  99ff6a83 Browse Directory
  
  Fangjun Kuang authored 2023-12-02 15:35:11 +0800
01 Dec, 2023 1 commit
- Fix CI (#456) · 539b27e5
  539b27e5 浏览文件
  
  Fangjun Kuang authored 2023-12-01 11:00:16 +0800
30 Nov, 2023 1 commit
- Use piper-phonemize to convert text to token IDs (#453) · 62dc3c3e
  62dc3c3e 浏览文件
  
  Fangjun Kuang authored 2023-11-30 23:57:43 +0800
28 Nov, 2023 1 commit
- Support piper-phonemize (#452) · db41778e
  db41778e 浏览文件
  
  Fangjun Kuang authored 2023-11-28 19:12:58 +0800
27 Nov, 2023 1 commit
- Release GIL to support multithreading in websocket servers. (#451) · 87a47d7d
  87a47d7d 浏览文件
  
  Fangjun Kuang authored 2023-11-27 13:44:03 +0800