Commit Graph

575 Commits

Author SHA1 Message Date
Georgi Gerganov
9c61f5f585
release : v1.4.1 2023-04-30 22:57:42 +03:00
Georgi Gerganov
c94c469592
whisper : fix quantize bug (#842)
* whisper : debug

* whisper : fix bug during quantization
2023-04-30 22:50:04 +03:00
Georgi Gerganov
feac80dd3f
ggml : fix UB (int << 31) 2023-04-30 22:27:30 +03:00
Georgi Gerganov
fa8dbdc888
release : v1.4.0 2023-04-30 19:23:37 +03:00
Georgi Gerganov
4a7d49af95
examples : fix + refactor Levenshtein distance 2023-04-30 19:12:49 +03:00
Georgi Gerganov
794b162a46
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
2023-04-30 18:51:57 +03:00
Georgi Gerganov
5fd1bdd7fc
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
2023-04-30 12:14:33 +03:00
Georgi Gerganov
0ccd6746c9
ggml : fix WASM build 2023-04-29 21:37:23 +03:00
Georgi Gerganov
d9b550c0a1
ggml : fix 32-bit ARM NEON (#836)
* ggml : add support for 32-bit ARM

* ggml : fix

* ggml : fix
2023-04-29 21:33:33 +03:00
Georgi Gerganov
e9b091c92a
ggml : use vzip instead of vuzp for consistency 2023-04-29 21:14:09 +03:00
Georgi Gerganov
1f30b99208
ggml : fix WASM build 2023-04-29 20:21:25 +03:00
Georgi Gerganov
05c3ea3bc8
ggml : sync with ggml repo (warning fixes + asserts) 2023-04-29 19:33:28 +03:00
Thijs Raymakers
6108d3cc58
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 <thijs@raymakers.nl>
2023-04-29 18:55:37 +03:00
Georgi Gerganov
bab97c83d0
tests : add "threads" to run-tests.sh 2023-04-29 12:32:28 +03:00
Georgi Gerganov
3eaeb030ff
extra : add sync-ggml.sh script 2023-04-29 12:32:28 +03:00
Georgi Gerganov
acec73ab6e
ggml : sync latest ggml + llama.cpp updates (quantization) 2023-04-29 12:32:28 +03:00
Zollner
5cc17418c7
whisper.android : add some tips (#816) 2023-04-29 11:00:20 +03:00
Georgi Gerganov
3efb81dec6
build : add WHISPER_COREML_ALLOW_FALLBACK to make / CMake (#812) 2023-04-29 10:55:24 +03:00
Canis Lupus
94a7cd2a07
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.
2023-04-29 10:49:02 +03:00
Georgi Gerganov
3e82ff4747
whisper : fix bug from previous commit 2023-04-29 10:42:14 +03:00
Georgi Gerganov
b5bd2f43c5
whisper : avoid designated initializers 2023-04-29 10:36:50 +03:00
AsukaMinato
94aa56f19e
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
2023-04-29 10:06:25 +03:00
Georgi Gerganov
4d89ee2e59
readme : add logo 2023-04-28 22:41:29 +03:00
Laytan Laats
70567eff23
main : escape quotes in csv output (#815) 2023-04-23 19:01:59 +03:00
Taras Glek
02ec83c5d5
stream : flush upon finishing inference (#811) 2023-04-23 17:00:30 +03:00
Philipp Zabel
2bd4b8d577
examples : add missing #include <cstdint> (#798)
common.cpp uses uint8_t and uint64_t, which are defined in <cstdint>.
2023-04-23 16:52:52 +03:00
Tauseef Mohiuddin
eecf2c3d41
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
2023-04-23 16:47:30 +03:00
Georgi Gerganov
c23588cc4b
release : v1.3.0 2023-04-15 17:30:44 +03:00
Georgi Gerganov
5108b30e6d
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
2023-04-15 17:19:19 +03:00
Georgi Gerganov
f19e23fbd1
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
2023-04-15 16:12:55 +03:00
Jhen-Jie Hong
ea1f8a50d4
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
2023-04-15 14:21:58 +03:00
Georgi Gerganov
3dead611bb
whisper : slightly faster Log Mel computation + n-1 FFT threads (#568) 2023-04-15 14:18:46 +03:00
Georgi Gerganov
355da83690
readme : fix link 2023-04-15 13:30:36 +03:00
Georgi Gerganov
3e5c49e59a
readme : add usage instructions for Core ML 2023-04-15 13:30:07 +03:00
Georgi Gerganov
5e47e223bd
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
2023-04-15 13:21:27 +03:00
Maximiliano Levi
794ff3074a
whisper : do not launch log_mel threads when n_thread is 1 (#763) 2023-04-14 22:35:34 +03:00
AfryMask
7e2afa4384
whisper : fix the bug related to word splitting errors in the "tokenize" function. (#760)
Co-authored-by: AfryMask <afrymask@gmail.com>
2023-04-14 20:35:03 +03:00
Aaron Taylor
1c5edc3cb3
readme : add SwiftWhisper to listed bindings (#755) 2023-04-14 20:24:00 +03:00
Georgi Gerganov
34b772727d
gitignore : add .test 2023-04-14 20:13:47 +03:00
Bader-eddine Ouaich
2c856fb9e5
whisper : fix potential memory leaks (#740)
* fix potential memory leak if whisper_init_state failed

* fix potential memory leak if gpt2_init failed
2023-04-14 20:05:56 +03:00
Anton Kostin
7727a40dc9
license : update year (#739) 2023-04-14 20:04:42 +03:00
GitAritron
b5639ed313
whisper : fix typos in whisper.h (#737)
Fixed a couple of typos (in comments, so nothing major). Keep up the great work 😄
2023-04-14 20:03:16 +03:00
Ali Alameh
2c4ac2627d
stream : support language auto-detect (#501)
#445  fix Language auto-detect "auto" flag does not work using the stream tool
2023-04-14 20:02:18 +03:00
Alex Evgrashin
674a8e579b
readme : add unity bindings (#733) 2023-04-14 19:59:44 +03:00
DGdev91
001083a769
talk, talk-llama : add basic example script for eleven-labs tts (#728) 2023-04-14 19:53:58 +03:00
Ivan Gorin
62b51c3070
models : change convert-pt-to-ggml to use .tiktoken tokenizer files (#725) 2023-04-14 19:50:39 +03:00
LittleLoli
61128870b8
cmake : add msvc compiler args /utf-8 fix error C3688 (#721)
* force msvc compiler use utf-8 encode

* only enable on msvc
2023-04-14 19:36:38 +03:00
Maciek
78548dc03f
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.
2023-04-14 19:36:09 +03:00
LittleLoli
66110dafcc
main : add lrc output support (#718)
* add lrc output support.

* fix wrong comment
2023-04-14 19:35:33 +03:00
Sam
b73a4638ac
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.
2023-04-14 19:33:06 +03:00