]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/log
pkg/ggml/sources/whisper.cpp
2 years agoggml : fix UB (int << 31)
Georgi Gerganov [Sun, 30 Apr 2023 19:27:30 +0000 (22:27 +0300)]
ggml : fix UB (int << 31)

2 years agorelease : v1.4.0
Georgi Gerganov [Sun, 30 Apr 2023 16:23:37 +0000 (19:23 +0300)]
release : v1.4.0

2 years agoexamples : fix + refactor Levenshtein distance
Georgi Gerganov [Sun, 30 Apr 2023 16:12:49 +0000 (19:12 +0300)]
examples : fix + refactor Levenshtein distance

2 years agowhisper : add integer quantization support (#540)
Georgi Gerganov [Sun, 30 Apr 2023 15:51:57 +0000 (18:51 +0300)]
whisper : add integer quantization support (#540)

* whisper : add integer quantization support

* examples : add common-ggml + prepare to add "quantize" tool

* whisper : quantization tool ready

* whisper : fix F32 support

* whisper : try to fix shared lib linkage

* wasm : update quantized models to Q5

* bench.wasm : remove "medium" button

* bench.wasm : fix custom model button

* ggml : add Q5_0 and Q5_1 WASM SIMD

* wasm : add quantized models to all WASM examples

* wasm : bump DB version number to 2

* talk-llama : update example to latest llama.cpp

* node : increase test timeout to 10s

* readme : add information for model quantization

* wasm : add links to other examples

2 years agowhisper : add GPU support via cuBLAS (#834)
Georgi Gerganov [Sun, 30 Apr 2023 09:14:33 +0000 (12:14 +0300)]
whisper : add GPU support via cuBLAS (#834)

* make : add WHISPER_CUBLAS

* make : fix CUBLAS build

* whisper : disable Flash Attention + adjust memory buffers

* whisper : remove old commented code

* readme : add cuBLAS instructions

* cmake : add WHISPER_CUBLAS option

* gitignore : ignore build-cublas

2 years agoggml : fix WASM build
Georgi Gerganov [Sat, 29 Apr 2023 18:37:23 +0000 (21:37 +0300)]
ggml : fix WASM build

2 years agoggml : fix 32-bit ARM NEON (#836)
Georgi Gerganov [Sat, 29 Apr 2023 18:33:33 +0000 (21:33 +0300)]
ggml : fix 32-bit ARM NEON (#836)

* ggml : add support for 32-bit ARM

* ggml : fix

* ggml : fix

2 years agoggml : use vzip instead of vuzp for consistency
Georgi Gerganov [Sat, 29 Apr 2023 18:14:09 +0000 (21:14 +0300)]
ggml : use vzip instead of vuzp for consistency

2 years agoggml : fix WASM build
Georgi Gerganov [Sat, 29 Apr 2023 17:21:25 +0000 (20:21 +0300)]
ggml : fix WASM build

2 years agoggml : sync with ggml repo (warning fixes + asserts)
Georgi Gerganov [Sat, 29 Apr 2023 16:30:22 +0000 (19:30 +0300)]
ggml : sync with ggml repo (warning fixes + asserts)

2 years agowhisper : use correct seek_end when offset is used (#833)
Thijs Raymakers [Sat, 29 Apr 2023 15:55:37 +0000 (17:55 +0200)]
whisper : use correct seek_end when offset is used (#833)

Whenever an `offset_ms` is provided, the value of `seek_end` is
calculated incorrectly. This causes Whisper to keep transcribing
after the end of the file.

The current behavior looks like
```
[00:34:40.000 --> 00:34:47.000]   This is an example audio file.
[00:34:47.000 --> 00:34:49.000]   The text has been redacted
[00:34:49.000 --> 00:34:51.000]   This is the end of the audio.
[00:34:51.000 --> 00:34:52.000]   ***
[00:34:52.000 --> 00:34:53.000]   ***
[00:34:53.000 --> 00:34:54.000]   ***
[00:34:55.000 --> 00:34:56.000]   ***
...
```

The expected behavior should be
```
[00:34:40.000 --> 00:34:47.000]   This is an example audio file.
[00:34:47.000 --> 00:34:49.000]   The text has been redacted
[00:34:49.000 --> 00:34:51.000]   This is the end of the audio.
- end of program -
```

This commit changes the calculation of the `seek_end` variable to
only add `seek_start` if a custom `duration_ms` is provided.
Otherwise, it defaults to the end of the file.

Signed-off-by: Thijs Raymakers <redacted>
2 years agotests : add "threads" to run-tests.sh
Georgi Gerganov [Sat, 29 Apr 2023 09:32:18 +0000 (12:32 +0300)]
tests : add "threads" to run-tests.sh

2 years agoextra : add sync-ggml.sh script
Georgi Gerganov [Sat, 29 Apr 2023 09:32:05 +0000 (12:32 +0300)]
extra : add sync-ggml.sh script

2 years agoggml : sync latest ggml + llama.cpp updates (quantization)
Georgi Gerganov [Sat, 29 Apr 2023 09:31:52 +0000 (12:31 +0300)]
ggml : sync latest ggml + llama.cpp updates (quantization)

2 years agowhisper.android : add some tips (#816)
Zollner [Sat, 29 Apr 2023 08:00:20 +0000 (16:00 +0800)]
whisper.android : add some tips (#816)

2 years agobuild : add WHISPER_COREML_ALLOW_FALLBACK to make / CMake (#812)
Georgi Gerganov [Sat, 29 Apr 2023 07:55:24 +0000 (10:55 +0300)]
build : add WHISPER_COREML_ALLOW_FALLBACK to make / CMake (#812)

2 years agowhisper : allow non-CoreML fallback when Core ML cannot be loaded (#812)
Canis Lupus [Sat, 29 Apr 2023 07:49:02 +0000 (08:49 +0100)]
whisper : allow non-CoreML fallback when Core ML cannot be loaded (#812)

if the Core ML model cannot be loaded, continue without Core ML instead of
returning. This allows a single build to transcribe using Core ML models
where available, and regular models when not.

2 years agowhisper : fix bug from previous commit
Georgi Gerganov [Sat, 29 Apr 2023 07:42:14 +0000 (10:42 +0300)]
whisper : fix bug from previous commit

2 years agowhisper : avoid designated initializers
Georgi Gerganov [Sat, 29 Apr 2023 07:36:50 +0000 (10:36 +0300)]
whisper : avoid designated initializers

2 years agominor : improve C++ and Python style (#768)
AsukaMinato [Sat, 29 Apr 2023 07:06:25 +0000 (16:06 +0900)]
minor : improve C++ and Python style (#768)

* use some STL functions

* use self.field than setattr, use pathlib.Path

* recover some format

* const some iter

* Keep the original

* 2 space

2 years agoreadme : add logo
Georgi Gerganov [Fri, 28 Apr 2023 19:41:29 +0000 (22:41 +0300)]
readme : add logo

2 years agomain : escape quotes in csv output (#815)
Laytan Laats [Sun, 23 Apr 2023 16:01:59 +0000 (18:01 +0200)]
main : escape quotes in csv output (#815)

2 years agostream : flush upon finishing inference (#811)
Taras Glek [Sun, 23 Apr 2023 14:00:30 +0000 (17:00 +0300)]
stream : flush upon finishing inference (#811)

2 years agoexamples : add missing #include <cstdint> (#798)
Philipp Zabel [Sun, 23 Apr 2023 13:52:52 +0000 (15:52 +0200)]
examples : add missing #include <cstdint> (#798)

common.cpp uses uint8_t and uint64_t, which are defined in <cstdint>.

2 years agomain : update escape_double_quotes() function (#776)
Tauseef Mohiuddin [Sun, 23 Apr 2023 13:47:30 +0000 (08:47 -0500)]
main : update escape_double_quotes() function (#776)

Updated the escape_double_quotes() function such that the function now escapes both double quotes and backslashes in the input string.

Changes Made:

- Renamed the function to escape_quotes_and_backslashes

- Modified the condition in the first loop to increment the value of 'escaped_length' for both double quotes and backslashes.

- Modified the condition in second loop to add a backslash before the current character if it is a double quote or a backslash.

Resolves: #769

2 years agorelease : v1.3.0
Georgi Gerganov [Sat, 15 Apr 2023 14:30:44 +0000 (17:30 +0300)]
release : v1.3.0

2 years agowhisper : pad audio instead of spectrogram (#579)
Georgi Gerganov [Sat, 15 Apr 2023 14:18:43 +0000 (17:18 +0300)]
whisper : pad audio instead of spectrogram (#579)

Also, fallback only if more temperatures are available and if we are
at least 3 seconds before the end of the audio

2 years agowhisper : restore decoder temperature fallbacks
Georgi Gerganov [Sat, 15 Apr 2023 13:04:07 +0000 (16:04 +0300)]
whisper : restore decoder temperature fallbacks

I disabled this because there were many complaints about slow decoding.
The current implementation does not allow batching the decoders when
using the "best of" or "beam size" parameters, so the decoding time is
proportional to the number of decoders, which is obviously not great.

However, now there are even more complaints about wrong decodings and
repetition.

So, making a compromise by re-enabling the fallbacks, but defaulting to
just 2 "best of" / "beam size" decoders. Also, the temperature step is
increased from 0.2 to 0.4 - i.e. from maximum of 5 fallbacks to maximum
of 2.

Also, the stream example now has fallbacks enabled by default.

close #471 #477 #508 #612 #719 #731

2 years agoggml, ci : fix build on whisper.android (ARM_NEON) + add CI (#764)
Jhen-Jie Hong [Sat, 15 Apr 2023 11:21:58 +0000 (19:21 +0800)]
ggml, ci : fix build on whisper.android (ARM_NEON) + add CI (#764)

* ggml : fix undefined symbol by remove inline handle

* ggml : make own ggml_aligned_malloc function

* ci: add ios/android build

2 years agowhisper : slightly faster Log Mel computation + n-1 FFT threads (#568)
Georgi Gerganov [Sat, 15 Apr 2023 11:18:46 +0000 (14:18 +0300)]
whisper : slightly faster Log Mel computation + n-1 FFT threads (#568)

2 years agoreadme : fix link
Georgi Gerganov [Sat, 15 Apr 2023 10:30:36 +0000 (13:30 +0300)]
readme : fix link

2 years agoreadme : add usage instructions for Core ML
Georgi Gerganov [Sat, 15 Apr 2023 10:30:07 +0000 (13:30 +0300)]
readme : add usage instructions for Core ML

2 years agowhisper : add Core ML support (#566)
Georgi Gerganov [Sat, 15 Apr 2023 10:21:27 +0000 (13:21 +0300)]
whisper : add Core ML support (#566)

* coreml : use Core ML encoder inference

* coreml : simlpify whisper_encode + log messages

* whisper : resolve rebase conflicts

* coreml : add scripts for CoreML model generation

* bench-all : recognize COREML flag

2 years agowhisper : do not launch log_mel threads when n_thread is 1 (#763)
Maximiliano Levi [Fri, 14 Apr 2023 19:35:34 +0000 (16:35 -0300)]
whisper : do not launch log_mel threads when n_thread is 1 (#763)

2 years agowhisper : fix the bug related to word splitting errors in the "tokenize" function...
AfryMask [Fri, 14 Apr 2023 17:35:03 +0000 (01:35 +0800)]
whisper : fix the bug related to word splitting errors in the "tokenize" function. (#760)

Co-authored-by: AfryMask <redacted>
2 years agoreadme : add SwiftWhisper to listed bindings (#755)
Aaron Taylor [Fri, 14 Apr 2023 17:24:00 +0000 (13:24 -0400)]
readme : add SwiftWhisper to listed bindings (#755)

2 years agogitignore : add .test
Georgi Gerganov [Fri, 14 Apr 2023 17:13:47 +0000 (20:13 +0300)]
gitignore : add .test

2 years agowhisper : fix potential memory leaks (#740)
Bader-eddine Ouaich [Fri, 14 Apr 2023 17:05:56 +0000 (17:05 +0000)]
whisper : fix potential memory leaks (#740)

* fix potential memory leak if whisper_init_state failed

* fix potential memory leak if gpt2_init failed

2 years agolicense : update year (#739)
Anton Kostin [Fri, 14 Apr 2023 17:04:42 +0000 (00:04 +0700)]
license : update year (#739)

2 years agowhisper : fix typos in whisper.h (#737)
GitAritron [Fri, 14 Apr 2023 17:03:16 +0000 (20:03 +0300)]
whisper : fix typos in whisper.h (#737)

Fixed a couple of typos (in comments, so nothing major). Keep up the great work ðŸ˜„

2 years agostream : support language auto-detect (#501)
Ali Alameh [Fri, 14 Apr 2023 17:02:18 +0000 (20:02 +0300)]
stream : support language auto-detect (#501)

#445  fix Language auto-detect "auto" flag does not work using the stream tool

2 years agoreadme : add unity bindings (#733)
Alex Evgrashin [Fri, 14 Apr 2023 16:59:44 +0000 (19:59 +0300)]
readme : add unity bindings (#733)

2 years agotalk, talk-llama : add basic example script for eleven-labs tts (#728)
DGdev91 [Fri, 14 Apr 2023 16:53:58 +0000 (18:53 +0200)]
talk, talk-llama : add basic example script for eleven-labs tts (#728)

2 years agomodels : change convert-pt-to-ggml to use .tiktoken tokenizer files (#725)
Ivan Gorin [Fri, 14 Apr 2023 16:50:39 +0000 (19:50 +0300)]
models : change convert-pt-to-ggml to use .tiktoken tokenizer files (#725)

2 years agocmake : add msvc compiler args /utf-8 fix error C3688 (#721)
LittleLoli [Fri, 14 Apr 2023 16:36:38 +0000 (00:36 +0800)]
cmake : add msvc compiler args /utf-8 fix error C3688 (#721)

* force msvc compiler use utf-8 encode

* only enable on msvc

2 years agotalk-llama : correct default speak.sh path (#720)
Maciek [Fri, 14 Apr 2023 16:36:09 +0000 (18:36 +0200)]
talk-llama : correct default speak.sh path (#720)

There is `speak.sh` file in `./examples/talk-llama` as described in README.
However `./examples/talk/speak.sh` is used in `talk-llama.cpp`, this commit corrects that.

2 years agomain : add lrc output support (#718)
LittleLoli [Fri, 14 Apr 2023 16:35:33 +0000 (00:35 +0800)]
main : add lrc output support (#718)

* add lrc output support.

* fix wrong comment

2 years agoreadme : make the quick start instructions clearer. (#716)
Sam [Fri, 14 Apr 2023 16:33:06 +0000 (16:33 +0000)]
readme : make the quick start instructions clearer. (#716)

Users wanting to make use of this implementation of the whisper model with no prior knowledge of C/C++ may download the Whisper model but fail to use of the "make" command as specified given that they forgot or didn't know they needed to clone the repository first. Hope this modification clears things up.

2 years agomake : disable avx in case f16c is not available (#706)
duthils [Fri, 14 Apr 2023 16:31:51 +0000 (12:31 -0400)]
make : disable avx in case f16c is not available (#706)

Why:

* ggml.c does not support AVX without F16C

2 years agoreadme : add shell command example for --print-colors (#710)
bocytko [Fri, 14 Apr 2023 16:25:23 +0000 (18:25 +0200)]
readme : add shell command example for --print-colors (#710)

The section of the readme file explaining `--print-colors` includes only a screenshot with directories that are inconsistent with other examples. This commit adds an example shell command, consistent with the remaining examples.

2 years agoggml : sync latest ggml
Georgi Gerganov [Fri, 14 Apr 2023 16:20:39 +0000 (19:20 +0300)]
ggml : sync latest ggml

2 years agowhisper : fix bug in prompt processing (close #705)
Georgi Gerganov [Fri, 14 Apr 2023 16:16:34 +0000 (19:16 +0300)]
whisper : fix bug in prompt processing (close #705)

Was dereferencing a dangling pointer

2 years agogo : exposed various parts to the Go Interface (#697)
Brian Murray [Fri, 14 Apr 2023 15:52:10 +0000 (09:52 -0600)]
go : exposed various parts to the Go Interface (#697)

2 years agoggml : fix q4_1 dot product types (#759)
novag [Fri, 14 Apr 2023 10:34:20 +0000 (12:34 +0200)]
ggml : fix q4_1 dot product types (#759)

Co-authored-by: Georgi Gerganov <redacted>
2 years agoggml : sync latest changes from ggml and llama.cpp
Georgi Gerganov [Thu, 13 Apr 2023 15:53:44 +0000 (18:53 +0300)]
ggml : sync latest changes from ggml and llama.cpp

2 years agoggml : fix WASM build
Georgi Gerganov [Mon, 10 Apr 2023 20:18:29 +0000 (23:18 +0300)]
ggml : fix WASM build

2 years agotalk-llama : increase context to 2048
Georgi Gerganov [Mon, 10 Apr 2023 20:09:15 +0000 (23:09 +0300)]
talk-llama : increase context to 2048

2 years agotalk-llama : update to latest llama.cpp (improved performance)
Georgi Gerganov [Mon, 10 Apr 2023 19:59:13 +0000 (22:59 +0300)]
talk-llama : update to latest llama.cpp (improved performance)

2 years agoggml : backport llama.cpp updates (close #709)
Georgi Gerganov [Mon, 10 Apr 2023 19:28:54 +0000 (22:28 +0300)]
ggml : backport llama.cpp updates (close #709)

- About x2 overall performance improvement on Apple Silicon
- Results should now be the same for different number of threads (not
  tested)

2 years agowhisper : add progress callback (#600)
pajowu [Thu, 30 Mar 2023 17:29:29 +0000 (19:29 +0200)]
whisper : add progress callback (#600)

2 years agomisc : typo (#688)
Zigfrid Zvezdin [Thu, 30 Mar 2023 04:51:33 +0000 (01:51 -0300)]
misc : typo (#688)

2 years agotalk-llama : fixing usage message for talk-llama (#687)
InconsolableCellist [Wed, 29 Mar 2023 21:10:20 +0000 (15:10 -0600)]
talk-llama : fixing usage message for talk-llama (#687)

"-ml" instead of "-mg" for specifying the llama file

2 years agomain : add <cstring> header
Georgi Gerganov [Wed, 29 Mar 2023 20:59:45 +0000 (23:59 +0300)]
main : add <cstring> header

2 years agowhisper.addon : fixed test to new async implementation (#686)
Lucas Zanek [Wed, 29 Mar 2023 20:59:17 +0000 (17:59 -0300)]
whisper.addon : fixed test to new async implementation (#686)

* fixed blocking code on node addon

* modify the example to run async

* format

* added logic to see the whisper output

* added logic to see the whisper output

* removed extra function for more clean example

* fixed whisper test to new async implementation

2 years agomodels : handle spaces and special characters in shell script paths (#677)
be-next [Wed, 29 Mar 2023 20:38:33 +0000 (22:38 +0200)]
models : handle spaces and special characters in shell script paths (#677)

This commit modifies the `get_script_path` function to correctly handle
spaces and special characters in directory paths. The fix involves adding
double quotes around variables and commands where needed to ensure proper
parsing of paths with spaces and special characters.

2 years agomain : fix typo in JSON output (#648)
Egor Egorov [Wed, 29 Mar 2023 20:26:39 +0000 (23:26 +0300)]
main : fix typo in JSON output (#648)

* typo in JSON output

* fix double quotes in JSON output

2 years agowhisper : add initial_prompt param (#645)
Jhen-Jie Hong [Wed, 29 Mar 2023 20:23:23 +0000 (04:23 +0800)]
whisper : add initial_prompt param (#645)

2 years agomake : 32-bit ARM flags (#486)
clach04 [Wed, 29 Mar 2023 20:11:35 +0000 (13:11 -0700)]
make : 32-bit ARM flags (#486)

* issue #470 - working 32-bit ARM

* Update Makefile

* Update Makefile

---------

Co-authored-by: Georgi Gerganov <redacted>
2 years agowhisper.swiftui : update README.md (#682)
Jonno [Wed, 29 Mar 2023 20:04:38 +0000 (06:04 +1000)]
whisper.swiftui : update README.md (#682)

- Slight tweaks to README for improved comprehension.

2 years agotalk-llama : add alpaca support (#668)
Evan Jones [Wed, 29 Mar 2023 20:01:14 +0000 (16:01 -0400)]
talk-llama : add alpaca support (#668)

2 years agowhisper : bump "large" scratch buffer even mode (close #671)
Georgi Gerganov [Tue, 28 Mar 2023 07:50:49 +0000 (10:50 +0300)]
whisper : bump "large" scratch buffer even mode (close #671)

2 years agowhisper : increase scratch buffers after recent change (#671)
Georgi Gerganov [Tue, 28 Mar 2023 07:36:16 +0000 (10:36 +0300)]
whisper : increase scratch buffers after recent change (#671)

Should fix the error:

ggml_new_tensor_impl: not enough space in the scratch memory

2 years agotalk-llama : add discussion link
Georgi Gerganov [Tue, 28 Mar 2023 07:11:34 +0000 (10:11 +0300)]
talk-llama : add discussion link

2 years agotalk-llama : try to fix windows build ..
Georgi Gerganov [Mon, 27 Mar 2023 18:28:00 +0000 (21:28 +0300)]
talk-llama : try to fix windows build ..

2 years agoreadme : add talk-llama example to the table
Georgi Gerganov [Mon, 27 Mar 2023 18:02:35 +0000 (21:02 +0300)]
readme : add talk-llama example to the table

2 years agotalk-llama : add new example + sync ggml from llama.cpp (#664)
Georgi Gerganov [Mon, 27 Mar 2023 18:00:32 +0000 (21:00 +0300)]
talk-llama : add new example + sync ggml from llama.cpp (#664)

* talk-llama : talk with LLaMA AI

* talk.llama : disable EOS token

* talk-llama : add README instructions

* ggml : fix build in debug

2 years agowhisper : disable fallbacks until the performance is improved (#588)
Georgi Gerganov [Wed, 22 Mar 2023 20:34:39 +0000 (22:34 +0200)]
whisper : disable fallbacks until the performance is improved (#588)

2 years agocmake : add a flag to disable F16C (#628)
Andrew Huynh [Wed, 22 Mar 2023 20:30:40 +0000 (13:30 -0700)]
cmake : add a flag to disable F16C (#628)

2 years agoInclude link to R wrapper in README (#626)
jwijffels [Wed, 22 Mar 2023 20:28:22 +0000 (21:28 +0100)]
Include link to R wrapper in README (#626)

2 years agoNodejs Addon blocking main thread. Implemented Napi::AsyncWorker (#642)
Lucas Zanek [Wed, 22 Mar 2023 20:19:22 +0000 (17:19 -0300)]
Nodejs Addon blocking main thread. Implemented Napi::AsyncWorker (#642)

* fixed blocking code on node addon

* modify the example to run async

* format

* added logic to see the whisper output

* added logic to see the whisper output

* removed extra function for more clean example

2 years agowhisper.objc : add `-O3 -DNDEBUG` in release mode (#640)
Jhen-Jie Hong [Wed, 22 Mar 2023 20:16:04 +0000 (04:16 +0800)]
whisper.objc : add `-O3 -DNDEBUG` in release mode (#640)

2 years agofixed language auto-detection for state provided processing (#627)
sandrohanea [Wed, 22 Mar 2023 19:47:09 +0000 (20:47 +0100)]
fixed language auto-detection for state provided processing (#627)

Co-authored-by: Sandro Hanea <redacted>
2 years agoreadme : add react-native bindings (#619)
Jhen-Jie Hong [Wed, 22 Mar 2023 19:39:02 +0000 (03:39 +0800)]
readme : add react-native bindings (#619)

2 years agomain : provide option for creating JSON output (#615)
Leo Moll [Wed, 22 Mar 2023 19:37:36 +0000 (20:37 +0100)]
main : provide option for creating JSON output (#615)

* examples : provide option for exporting also as JSON file (ggerganov/whisper.cpp#614)

* main : remove leftovers

---------

Co-authored-by: Georgi Gerganov <redacted>
2 years agomodels : change default encoding to utf8 (#605)
Kamilake [Wed, 22 Mar 2023 19:17:24 +0000 (04:17 +0900)]
models : change default encoding to utf8 (#605)

2 years agomake : fix MUSL Linux build (#576)
Georgi Gerganov [Wed, 22 Mar 2023 18:51:42 +0000 (20:51 +0200)]
make : fix MUSL Linux build (#576)

2 years agomodels : change HF hosting from dataset to model
Georgi Gerganov [Wed, 22 Mar 2023 18:44:56 +0000 (20:44 +0200)]
models : change HF hosting from dataset to model

2 years agowhisper.android : support benchmark for Android example. (#542)
Takeshi Inoue [Tue, 7 Mar 2023 19:36:30 +0000 (04:36 +0900)]
whisper.android : support benchmark for Android example. (#542)

* whisper.android: Support benchmark for Android example.

* whisper.android: update screenshot in README.

* update: Make text selectable for copy & paste.

* Update whisper.h to restore API name

Co-authored-by: Georgi Gerganov <redacted>
* whisper.android: Restore original API names.

---------

Co-authored-by: tinoue <redacted>
Co-authored-by: Georgi Gerganov <redacted>
2 years agoreadme : add bench-wts.sh demo
Georgi Gerganov [Mon, 6 Mar 2023 19:06:27 +0000 (21:06 +0200)]
readme : add bench-wts.sh demo

2 years agobench-wts.sh : rename script + add execute permission
Georgi Gerganov [Mon, 6 Mar 2023 19:02:24 +0000 (21:02 +0200)]
bench-wts.sh : rename script + add execute permission

2 years agoqual-bench.sh : add quality comparison tool, and update main.cpp to allow using a...
venkr [Mon, 6 Mar 2023 17:18:11 +0000 (09:18 -0800)]
qual-bench.sh : add quality comparison tool, and update main.cpp to allow using a font file (#569)

2 years agowhisper.android : enable fp16 instrinsics (FP16_VA) which is supported by ARMv8.2...
Takeshi Inoue [Mon, 6 Mar 2023 17:15:57 +0000 (02:15 +0900)]
whisper.android : enable fp16 instrinsics (FP16_VA) which is supported by ARMv8.2 or later. (#572)

2 years agowhisper : add whisper_state + default state on the whisper_context (#523)
sandrohanea [Sun, 5 Mar 2023 19:42:19 +0000 (20:42 +0100)]
whisper : add whisper_state + default state on the whisper_context (#523)

* Added whisper state + default state on the whisper_context

* Fixed some examples and bindings

* Fixed whisper_n_len (which was used in some binding) and added whisper_n_len_from_state

* Fixed comments

* whisper : reuse kv_cache_free() and fix compiler warnings

* whisper : clean-up the API comments

---------

Co-authored-by: Sandro Hanea <redacted>
Co-authored-by: Georgi Gerganov <redacted>
2 years agowhisper : set no_context == true by default (#537)
Georgi Gerganov [Sun, 5 Mar 2023 18:53:43 +0000 (20:53 +0200)]
whisper : set no_context == true by default (#537)

2 years agogo : NewContext now returns a clean context (#537)
polarmoon [Sun, 5 Mar 2023 18:50:25 +0000 (10:50 -0800)]
go : NewContext now returns a clean context (#537)

Co-authored-by: Ming <redacted>
2 years agomain : add csv header (#552)
HY. Kelvin Lee [Thu, 2 Mar 2023 16:32:16 +0000 (11:32 -0500)]
main : add csv header (#552)

2 years agomake : add -DNDEBUG compile flag
Georgi Gerganov [Tue, 28 Feb 2023 21:27:54 +0000 (23:27 +0200)]
make : add -DNDEBUG compile flag

2 years agorelease : v1.2.1
Georgi Gerganov [Tue, 28 Feb 2023 20:29:12 +0000 (22:29 +0200)]
release : v1.2.1

2 years agomake : add "-mcpu=native" when building for aarch64 (#532)
FlippFuzz [Mon, 27 Feb 2023 19:04:16 +0000 (03:04 +0800)]
make : add "-mcpu=native" when building for aarch64 (#532)

2 years agoreadme : add pybind11 bindings (#538)
Aaron Pham [Mon, 27 Feb 2023 19:02:11 +0000 (11:02 -0800)]
readme : add pybind11 bindings (#538)