Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Bug]: Whisper crash discussions #370

Open
thewh1teagle opened this issue Nov 11, 2024 · 6 comments
Open

[Bug]: Whisper crash discussions #370

thewh1teagle opened this issue Nov 11, 2024 · 6 comments
Labels
bug Something isn't working

Comments

@thewh1teagle
Copy link
Owner

thewh1teagle commented Nov 11, 2024

Whisper crash discussions

Instead of creating multiple issues, here we'll talk about whisper crashes (mostly on Windows)
Once it solved for you, you can disable the notifications for this thread.

The first helpful thing you can do:

Open cmd.exe (not powershell)
Run this, try to transcribe. copy logs and paste here :)

taskkill /IM vibe.exe /F
set RUST_BACKTRACE=1
set RUST_LOG=debug
%localappdata%\vibe\vibe.exe

macOS:

RUST_LOG=debug /Applications/vibe.app/Contents/MacOS/vibe

Linux

RUST_LOG=debug vibe

Other things you can try:

@forai108
Copy link

forai108 commented Nov 11, 2024

Log

Microsoft Windows [Version 10.0.19044.1288] (c) Microsoft Corporation. All rights reserved.

C:\Users\Wael>taskkill /IM vibe.exe /F
ERROR: The process "vibe.exe" not found.

C:\Users\Wael>set RUST_BACKTRACE=1

C:\Users\Wael>set RUST_LOG=debug

C:\Users\Wael>%localappdata%\vibe\vibe.exe

C:\Users\Wael>2024-11-11T03:08:46.420618Z DEBUG vibe::setup: Vibe App Running
2024-11-11T03:08:46.422950Z DEBUG vibe::setup: webview version: 130.0.2849.56
2024-11-11T03:08:46.423208Z DEBUG vibe::custom_protocol: Protocol handler registered successfully.
2024-11-11T03:08:46.423449Z DEBUG vibe::setup: Cargo features: vulkan
2024-11-11T03:08:46.423697Z DEBUG vibe::setup: CPU Features
{"avx":{"enabled":true,"support":true},"avx2":{"enabled":true,"support":true},"f16c":{"enabled":true,"support":true},"fma":{"enabled":true,"support":true}}
2024-11-11T03:08:46.424000Z DEBUG vibe::setup: Executable Architecture: x86_64
2024-11-11T03:08:46.424237Z DEBUG vibe::setup: APP VERSION: 2.6.5
2024-11-11T03:08:46.424510Z DEBUG vibe::setup: COMMIT HASH: c2be67d
2024-11-11T03:08:46.424725Z DEBUG vibe::setup: Non CLI mode
2024-11-11T03:08:47.245189Z DEBUG vibe::cmd: Vulkan support is successfully checked and working.
2024-11-11T03:08:47.250620Z DEBUG resolve{host=github.com}: hyper_util::client::legacy::connect::dns: resolving host="github.com"
2024-11-11T03:08:47.259455Z DEBUG vibe::cmd::audio: Default Input Device:
Ok("Microphone Array (Realtek(R) Audio)")
2024-11-11T03:08:47.260063Z DEBUG vibe::cmd::audio: Default Output Device:
Ok("Speaker/Headphone (Realtek(R) Audio)")
2024-11-11T03:08:47.261386Z DEBUG vibe::cmd::audio: Devices:
2024-11-11T03:08:47.302035Z DEBUG hyper_util::client::legacy::connect::http: connecting to 140.82.121.3:443
2024-11-11T03:08:47.374087Z DEBUG hyper_util::client::legacy::connect::http: connected to 140.82.121.3:443
2024-11-11T03:08:47.713565Z DEBUG hyper_util::client::legacy::pool: pooling idle connection for ("https", github.com)
2024-11-11T03:08:47.714192Z DEBUG hyper_util::client::legacy::pool: reuse idle connection for ("https", github.com)
2024-11-11T03:08:47.921643Z DEBUG hyper_util::client::legacy::pool: pooling idle connection for ("https", github.com)
2024-11-11T03:08:47.922201Z DEBUG resolve{host=objects.githubusercontent.com}: hyper_util::client::legacy::connect::dns: resolving host="objects.githubusercontent.com"
2024-11-11T03:08:47.924130Z DEBUG hyper_util::client::legacy::connect::http: connecting to 185.199.111.133:443
2024-11-11T03:08:47.988162Z DEBUG hyper_util::client::legacy::connect::http: connected to 185.199.111.133:443
2024-11-11T03:08:52.846303Z DEBUG vibe::cmd: loading model first time
2024-11-11T03:08:52.846565Z DEBUG vibe_core::transcribe: open model...
2024-11-11T03:08:52.847755Z DEBUG vibe_core::transcribe: gpu device: 0
2024-11-11T03:08:52.851189Z DEBUG vibe_core::transcribe: use gpu: true
2024-11-11T03:08:52.851835Z DEBUG vibe_core::transcribe: creating whisper context with model path C:\Users\Wael\AppData\Local\github.com.thewh1teagle.vibe\ggml-large-v1.bin
2024-11-11T03:08:52.852577Z INFO whisper_rs::whisper_sys_tracing: whisper_init_from_file_with_params_no_state: loading model from 'C:\Users\Wael\AppData\Local\github.com.thewh1teagle.vibe\ggml-large-v1.bin'
2024-11-11T03:08:52.853629Z INFO whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: use gpu = 1
2024-11-11T03:08:52.853986Z INFO whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: flash attn = 0
2024-11-11T03:08:52.854661Z INFO whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: gpu_device = 0
2024-11-11T03:08:52.855240Z INFO whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: dtw = 0
2024-11-11T03:08:52.917186Z DEBUG whisper_rs::whisper_sys_tracing: ggml_vulkan: Found 1 Vulkan devices:
2024-11-11T03:08:52.918953Z DEBUG whisper_rs::whisper_sys_tracing: ggml_vulkan: 0 = NVIDIA GeForce GTX 1050 with Max-Q Design (NVIDIA) | uma: 0 | fp16: 0 | warp size: 32
2024-11-11T03:08:52.919317Z DEBUG whisper_rs::whisper_sys_tracing: register_backend: registered backend Vulkan (1 devices)
2024-11-11T03:08:52.920687Z DEBUG whisper_rs::whisper_sys_tracing: register_device: registered device Vulkan0 (NVIDIA GeForce GTX 1050 with Max-Q Design)
2024-11-11T03:08:52.922884Z DEBUG whisper_rs::whisper_sys_tracing: register_backend: registered backend CPU (1 devices)
2024-11-11T03:08:52.924760Z DEBUG whisper_rs::whisper_sys_tracing: register_device: registered device CPU (Intel(R) Core(TM) i7-8565U CPU @ 1.80GHz)
2024-11-11T03:08:52.925341Z INFO whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: backends = 2
2024-11-11T03:08:52.926784Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: loading model
2024-11-11T03:08:52.927674Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_vocab = 51865
2024-11-11T03:08:52.928203Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_audio_ctx = 1500
2024-11-11T03:08:52.928746Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_audio_state = 1280
2024-11-11T03:08:52.929259Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_audio_head = 20
2024-11-11T03:08:52.930072Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_audio_layer = 32
2024-11-11T03:08:52.930470Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_text_ctx = 448
2024-11-11T03:08:52.930861Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_text_state = 1280
2024-11-11T03:08:52.931308Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_text_head = 20
2024-11-11T03:08:52.931699Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_text_layer = 32
2024-11-11T03:08:52.932212Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_mels = 80
2024-11-11T03:08:52.932685Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: ftype = 1
2024-11-11T03:08:52.933124Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: qntvr = 0
2024-11-11T03:08:52.933496Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: type = 5 (large)
2024-11-11T03:08:53.004239Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: adding 1608 extra tokens
2024-11-11T03:08:53.006464Z INFO whisper_rs::whisper_sys_tracing: whisper_model_load: n_langs = 99
ggml_vulkan: Device memory allocation of size 3093987840 failed.
ggml_vulkan: vk::Device::allocateMemory: ErrorOutOfDeviceMemory
2024-11-11T03:08:53.101435Z DEBUG whisper_rs::whisper_sys_tracing: alloc_tensor_range: failed to allocate Vulkan0 buffer of size 3093987840
2024-11-11T03:08:53.102090Z ERROR whisper_rs::whisper_sys_tracing: whisper_model_load: failed to allocate memory for the model
2024-11-11T03:08:53.102821Z ERROR whisper_rs::whisper_sys_tracing: whisper_init_with_params_no_state: failed to load model

@thewh1teagle
Copy link
Owner Author

Log

Vulkan says that your device out of memory. try with the default model that comes with vibe. change it to medium model and try again.

@forai108
Copy link

forai108 commented Nov 11, 2024

Yeah, the medium one has worked. Is the problem with vram or ram for the large model ?

@thewh1teagle
Copy link
Owner Author

Yeah, the medium one has worked. Is the problem with vram or ram for the large model ?

With the VRAM of the GPU. you don't have enough VRAM.

@forai108
Copy link

I understand, Thanks for your help.

@oleole39
Copy link
Contributor

I find that topic too late, so here's one in Linux opened as a separate issue: #395

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

3 participants