]>
git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/log
Georgi Gerganov [Fri, 25 Nov 2022 17:08:51 +0000 (19:08 +0200)]
refactoring : more readable code
vicalloy [Fri, 25 Nov 2022 03:24:08 +0000 (11:24 +0800)]
correct model name display on running samples
Georgi Gerganov [Thu, 24 Nov 2022 21:13:26 +0000 (23:13 +0200)]
wasm : refactor wasm example + reuse fetch mechanism
Georgi Gerganov [Thu, 24 Nov 2022 18:15:07 +0000 (20:15 +0200)]
talk.wasm : update video link + some minor fixes
Georgi Gerganov [Thu, 24 Nov 2022 18:09:45 +0000 (20:09 +0200)]
Update README.md
Use a less cringy video to demo talk.wasm lol
Georgi Gerganov [Thu, 24 Nov 2022 18:06:51 +0000 (20:06 +0200)]
Update README.md
Georgi Gerganov [Thu, 24 Nov 2022 16:24:06 +0000 (18:24 +0200)]
talk.wasm : move to https://whisper.ggerganov.com/talk
This way, we can share the same models across different WASM examples
and not have to download them for each page
Georgi Gerganov [Thu, 24 Nov 2022 15:54:41 +0000 (17:54 +0200)]
models : add instructions for using HF fine-tuned models
Georgi Gerganov [Thu, 24 Nov 2022 15:54:16 +0000 (17:54 +0200)]
whisper : improve printfs
Georgi Gerganov [Thu, 24 Nov 2022 15:53:51 +0000 (17:53 +0200)]
main : fix dangling pointer when using stdin for input (#65)
Georgi Gerganov [Thu, 24 Nov 2022 15:52:04 +0000 (17:52 +0200)]
main, stream : remove --verbose flag (#178)
Georgi Gerganov [Wed, 23 Nov 2022 22:34:00 +0000 (00:34 +0200)]
talk.wasm : add audio pre-processing + bump memory
Georgi Gerganov [Wed, 23 Nov 2022 22:08:57 +0000 (00:08 +0200)]
talk.wasm : refactoring + update README.md
Georgi Gerganov [Wed, 23 Nov 2022 21:22:40 +0000 (23:22 +0200)]
models : add usage comments to the HF convert script (#157)
Georgi Gerganov [Wed, 23 Nov 2022 21:14:11 +0000 (23:14 +0200)]
models : fix HF fine-tuned model conversion script (#157)
It works now
Georgi Gerganov [Wed, 23 Nov 2022 20:40:06 +0000 (22:40 +0200)]
ggml : fix the fix
Georgi Gerganov [Wed, 23 Nov 2022 20:27:49 +0000 (22:27 +0200)]
ggml : fix cross-compile Linux -> Window with mingw (#168)
Georgi Gerganov [Wed, 23 Nov 2022 20:16:50 +0000 (22:16 +0200)]
Revert "update README.md"
This reverts commit
6a84147113669bed68bbc4d31e3c14f914092bf8 .
katsu560 [Wed, 23 Nov 2022 13:59:54 +0000 (22:59 +0900)]
update README.md
katsu560 [Wed, 23 Nov 2022 13:54:21 +0000 (22:54 +0900)]
ggml: change inline ggml_fp16_to_fp32, ggml_fp16_t ggml_fp32_to_fp16
katsu560 [Wed, 23 Nov 2022 12:31:05 +0000 (21:31 +0900)]
add gprof option
katsu560 [Wed, 23 Nov 2022 11:23:35 +0000 (20:23 +0900)]
fix AVX,AVX2,FMA,F16C detection on Linux and add flags for OpenBLAS
katsu560 [Wed, 23 Nov 2022 11:23:24 +0000 (20:23 +0900)]
add AVX support
Tamotsu Takahashi [Wed, 23 Nov 2022 06:17:13 +0000 (15:17 +0900)]
Build with OpenBLAS and SDL2 on windows
Georgi Gerganov [Wed, 23 Nov 2022 20:07:20 +0000 (22:07 +0200)]
models : minor changes to the HF convert script (#157)
Georgi Gerganov [Wed, 23 Nov 2022 15:17:31 +0000 (17:17 +0200)]
models : add "convert-h5-to-ggml.py" script (#157)
Converts transformers models to ggml.
Although the conversion is successful, it does not work for some reason.
Not sure why
Georgi Gerganov [Wed, 23 Nov 2022 15:17:01 +0000 (17:17 +0200)]
minor : updates few prints + fix buttons in whisper.wasm
Georgi Gerganov [Wed, 23 Nov 2022 07:53:55 +0000 (09:53 +0200)]
Update README.md
Georgi Gerganov [Wed, 23 Nov 2022 07:52:36 +0000 (09:52 +0200)]
Update README.md
Tamotsu Takahashi [Wed, 23 Nov 2022 00:46:56 +0000 (09:46 +0900)]
Find libopenblas.dll.a on windows
"lib" is needed for windows.
With this change, you can build whisper.cpp with OpenBLAS's prebuilt DLL.
1. extract a zip from https://github.com/xianyi/OpenBLAS/releases
2. copy the headers in (openblas)/include to the root directory of whisper.cpp
3. invoke cmake with -DCMAKE_LIBRARY_PATH=(openblas)\lib -DWHISPER_SUPPORT_OPENBLAS=ON
4. copy (openblas)/bin/libopenblas.dll to the same directory of whisper.dll after msbuild
https://github.com/ggerganov/whisper.cpp/issues/89#issuecomment-
1324391258
Georgi Gerganov [Wed, 23 Nov 2022 06:24:29 +0000 (08:24 +0200)]
unicode : fix character replacement (thanks to @tamo)
Georgi Gerganov [Tue, 22 Nov 2022 20:48:56 +0000 (22:48 +0200)]
close #109 : add fetching of the model over HTTP (whisper.wasm)
Georgi Gerganov [Tue, 22 Nov 2022 20:22:17 +0000 (22:22 +0200)]
talk.wasm : final touches
Georgi Gerganov [Tue, 22 Nov 2022 18:10:20 +0000 (20:10 +0200)]
talk.wasm : polishing + adding many AI personalities
Georgi Gerganov [Tue, 22 Nov 2022 16:20:05 +0000 (18:20 +0200)]
stream : "-kc" now enables context keeping from previous segment (#90)
By default, the context keeping is disabled
M. Eren Akbiyik [Tue, 22 Nov 2022 16:10:35 +0000 (17:10 +0100)]
Prompt previous tokens for streaming (#163)
* feat: prompt previous tokens for streaming
I used a vector pointer instead of vector itself because it gave weird errors, and why not
* convert vector to use with C api
* feat: remove old refs, check for prompt size
* feat: use better way of getting the pointer
Georgi Gerganov [Mon, 21 Nov 2022 20:42:29 +0000 (22:42 +0200)]
talk.wasm : update README.md
Georgi Gerganov [Mon, 21 Nov 2022 20:20:42 +0000 (22:20 +0200)]
talk.wasm : GPT-2 meets Whisper in WebAssembly (#155)
* talk : initial real-time transcription in the browser
* talk : polishing the UI
* talk : ready for beta testing
* talk.wasm : rename example
Georgi Gerganov [Mon, 21 Nov 2022 16:52:20 +0000 (18:52 +0200)]
Update README.md
Georgi Gerganov [Sun, 20 Nov 2022 20:43:32 +0000 (22:43 +0200)]
ggml : fix Windows build
Georgi Gerganov [Sun, 20 Nov 2022 20:39:39 +0000 (22:39 +0200)]
ci : add Windows build
Georgi Gerganov [Sun, 20 Nov 2022 19:22:02 +0000 (21:22 +0200)]
stream : add "max_tokens" cli arg
Controls the max tokens per segment for the stream example
Georgi Gerganov [Sun, 20 Nov 2022 19:12:01 +0000 (21:12 +0200)]
stream : add "audio_ctx" parameter
Used to overwrite the audio context size of the Encoder.
For example, setting "audio_ctx = 512" will make it run about 3 times
faster, processing about 10s of audio, instead of 30s.
The transcription quality drops, but this can be used for real-time
streaming purposes where performance is important.
Georgi Gerganov [Sun, 20 Nov 2022 18:52:24 +0000 (20:52 +0200)]
stream : add "max_tokens" parameter
Used to limit the number of tokens in a segment.
Useful to battle with word repetition when using partial encoder context
Georgi Gerganov [Sun, 20 Nov 2022 18:45:10 +0000 (20:45 +0200)]
stream : add "single_segment" option
Force the entire audio chunk to be transcribed into a single segment
Georgi Gerganov [Fri, 11 Nov 2022 20:33:10 +0000 (22:33 +0200)]
stream : partial encoder experiments
greeshmay [Thu, 17 Nov 2022 20:12:51 +0000 (12:12 -0800)]
fix: free ggml_context (close #149) (#150)
* fix: free ggml_context
* ggml : free the model's contexts in whisper_free()
Co-authored-by: Georgi Gerganov <redacted>
Georgi Gerganov [Wed, 16 Nov 2022 17:21:43 +0000 (19:21 +0200)]
models : simplify the conversion script
"transformers" dependency is not actually needed
Dody Suria Wijaya [Wed, 16 Nov 2022 16:53:01 +0000 (23:53 +0700)]
Update download-ggml-model.sh
follow curl redirect to new hosting site
Georgi Gerganov [Tue, 15 Nov 2022 17:47:06 +0000 (19:47 +0200)]
models : change default hosting to Hugging Face
My Linode is running out of monthly bandwidth due to the big interest in
the project
Georgi Gerganov [Sat, 12 Nov 2022 16:03:49 +0000 (18:03 +0200)]
whisper : add option to speed up the audio tempo by x2
Using a Phase Vocoder for speeding up the audio tempo by scaling down
the frequencies in the frequency domain.
This reduces the computation in the Encoder by a factor of 2.
The transcription accuracy is degraded, but for slow to normal speech -
it seems to be still very good.
I think this can find application for real-time transcription - i.e. the
"stream" example.
Georgi Gerganov [Sun, 13 Nov 2022 07:08:33 +0000 (09:08 +0200)]
make : add libwhisper.so target (#144)
Chidi Williams [Fri, 11 Nov 2022 16:10:01 +0000 (16:10 +0000)]
Add WHISPER_NO_AVX and WHISPER_NO_AVX2 to CMakeLists (#136)
* Check for AVX and AVX2 on Darwin
* Add AVX options to CMakeLists
Georgi Gerganov [Fri, 11 Nov 2022 16:02:58 +0000 (18:02 +0200)]
minor : remove one more redundant line
Georgi Gerganov [Fri, 11 Nov 2022 15:58:51 +0000 (17:58 +0200)]
minor : fix double float32 conversion in python script
Georgi Gerganov [Wed, 9 Nov 2022 19:41:21 +0000 (21:41 +0200)]
ref #40 : start working on the documentation
Alan [Wed, 9 Nov 2022 18:24:06 +0000 (15:24 -0300)]
Adds support for stdin wav input
Georgi Gerganov [Wed, 9 Nov 2022 17:32:58 +0000 (19:32 +0200)]
js : update whipser.js to latest
Chidi Williams [Wed, 9 Nov 2022 00:28:36 +0000 (00:28 +0000)]
Check for AVX and AVX2 on Darwin
boolemancer [Tue, 8 Nov 2022 11:04:23 +0000 (03:04 -0800)]
Fix the Windows pthread_create shim
The current implementation doesn't actually set the out parameter,
and it returns 0 on failure instead of on success.
Georgi Gerganov [Mon, 7 Nov 2022 19:48:13 +0000 (21:48 +0200)]
sync : submodule whisper.spm
Georgi Gerganov [Mon, 7 Nov 2022 18:50:24 +0000 (20:50 +0200)]
cmake : add submodule whisper.spm
Georgi Gerganov [Mon, 7 Nov 2022 18:14:52 +0000 (20:14 +0200)]
ref #22 : add "duration" option
Can be used to partially process a recording
Georgi Gerganov [Sun, 6 Nov 2022 19:04:21 +0000 (21:04 +0200)]
Update README.md
Georgi Gerganov [Sun, 6 Nov 2022 07:22:50 +0000 (09:22 +0200)]
examples : add simple script for generating Karaoke video
Georgi Gerganov [Sat, 5 Nov 2022 06:44:41 +0000 (08:44 +0200)]
Update README.md
Georgi Gerganov [Fri, 4 Nov 2022 20:26:08 +0000 (22:26 +0200)]
Update README.md
Georgi Gerganov [Fri, 4 Nov 2022 16:30:38 +0000 (18:30 +0200)]
main : fix generated bash script
Georgi Gerganov [Thu, 3 Nov 2022 18:53:44 +0000 (20:53 +0200)]
ggml : multi-thread the ggml_add operator
Georgi Gerganov [Thu, 3 Nov 2022 18:18:57 +0000 (20:18 +0200)]
cmake : fix passing GGML_PERF compile option
Georgi Gerganov [Wed, 2 Nov 2022 20:03:27 +0000 (22:03 +0200)]
Update README.md
Georgi Gerganov [Wed, 2 Nov 2022 19:18:20 +0000 (21:18 +0200)]
whisper : token-level timestamp refactoring (#49, #120)
This turned out pretty good overall. The algorithm has been moved from
main.cpp to whisper.cpp and can be reused for all subtitles types. This
means that now you can specify the maximum length of the generated
lines. Simply provide the "-ml" argument specifying the max length in
number of characters
Georgi Gerganov [Wed, 2 Nov 2022 16:33:29 +0000 (18:33 +0200)]
Update README.md
Georgi Gerganov [Wed, 2 Nov 2022 16:31:55 +0000 (18:31 +0200)]
extra : compute SHA of all models files
Georgi Gerganov [Wed, 2 Nov 2022 16:31:18 +0000 (18:31 +0200)]
whisper : fix extra memory usage after recent processor changes
Had increased the memory buffer to the size of the model and forgot to
bring it down.
Syed Jafri [Wed, 2 Nov 2022 16:00:19 +0000 (10:00 -0600)]
Allow building with Accelerate for x86_64 Macs (#123)
* Cross compile windows
* set env properly
* rm log
* fix review
* Add back space
* Don't force architecture
* Allow building x86_64 with accelerate
Georgi Gerganov [Wed, 2 Nov 2022 15:52:24 +0000 (17:52 +0200)]
ggml : fix the check for NEON support (#7)
Was using the wrong preprocessor macro
Syed Jafri [Wed, 2 Nov 2022 06:46:49 +0000 (00:46 -0600)]
Cross compilation (#121)
* Cross compile windows
* set env properly
* rm log
* fix review
* Add back space
Georgi Gerganov [Tue, 1 Nov 2022 20:47:58 +0000 (22:47 +0200)]
Update README.md
Georgi Gerganov [Tue, 1 Nov 2022 20:35:21 +0000 (22:35 +0200)]
main : add some comments for the word-level timestamp algorithm
Georgi Gerganov [Tue, 1 Nov 2022 20:09:25 +0000 (22:09 +0200)]
main : fix some edge cases for word-level timestamps
Georgi Gerganov [Mon, 31 Oct 2022 20:06:05 +0000 (22:06 +0200)]
Update README.md
Georgi Gerganov [Mon, 31 Oct 2022 18:19:41 +0000 (20:19 +0200)]
Update README.md
Mikhail Grigorev [Sun, 30 Oct 2022 19:51:29 +0000 (00:51 +0500)]
Added for Windows implemenated script download-ggml-model.cmd
Mikhail Grigorev [Sun, 30 Oct 2022 17:19:24 +0000 (22:19 +0500)]
Fixed sched_yield
Mikhail Grigorev [Sun, 30 Oct 2022 10:29:27 +0000 (15:29 +0500)]
Implemenated sched_yield function for Windows
Georgi Gerganov [Sun, 30 Oct 2022 15:11:37 +0000 (17:11 +0200)]
Update README.md
Georgi Gerganov [Sun, 30 Oct 2022 15:10:46 +0000 (17:10 +0200)]
Update README.md
Georgi Gerganov [Sun, 30 Oct 2022 08:05:58 +0000 (10:05 +0200)]
main : add option for word-leve timestamps (very experimental)
Georgi Gerganov [Sun, 30 Oct 2022 06:27:04 +0000 (08:27 +0200)]
stream : add "--capture" option to select capture device (ref #10)
Georgi Gerganov [Sun, 30 Oct 2022 06:23:52 +0000 (08:23 +0200)]
close #113 : fix struct whisper_token_data
Georgi Gerganov [Sat, 29 Oct 2022 18:28:06 +0000 (21:28 +0300)]
minor : update whisper.js
Georgi Gerganov [Sat, 29 Oct 2022 17:30:05 +0000 (20:30 +0300)]
whisper.wasm : update system info print
Georgi Gerganov [Sat, 29 Oct 2022 16:41:50 +0000 (19:41 +0300)]
ref #5 : update CMake for Windows build
- __AVX2__ should already be defined due to /arch:AVX2
- _CRT_SECURE_NO_WARNINGS should be defined both for shared and static lib
Georgi Gerganov [Sat, 29 Oct 2022 11:14:23 +0000 (14:14 +0300)]
minor : fix multiple definitions of to_timestamp()
Georgi Gerganov [Sat, 29 Oct 2022 11:08:23 +0000 (14:08 +0300)]
parallel : print time of audio boundaries + fix timings
Georgi Gerganov [Sat, 29 Oct 2022 09:38:29 +0000 (12:38 +0300)]
ggml : fix barrier
Georgi Gerganov [Sat, 29 Oct 2022 09:26:03 +0000 (12:26 +0300)]
main : merge parallel example in main
Georgi Gerganov [Sat, 29 Oct 2022 09:24:02 +0000 (12:24 +0300)]
parallel : working
Georgi Gerganov [Sat, 29 Oct 2022 08:23:44 +0000 (11:23 +0300)]
ggml : fix thread-safety of ggml_init and ggml_free