search

chidiwilliams / buzz

Buzz transcribes and translates audio offline on your personal computer. Powered by OpenAI's Whisper.
https://chidiwilliams.github.io/buzz
MIT License
12.64k stars 948 forks source link

[Bug]Import URL feature broken #964

Closed mriamnobody closed 4 weeks ago

mriamnobody commented 4 weeks ago

Environment

OS: Windows 11 (26100.1882) Python: 3.10.7 CUDA Version: 12.1 ffmpeg version on path: 6.1.1

The Import URL feature is not functioning as expected. I attempted using various models with this feature, but the issue persists across all cases.

https://github.com/user-attachments/assets/0b19e02f-d26a-4731-b850-d00a086583d8

[2024-10-27 06:49:32,015] backend._load_plugins:205 DEBUG -> Loading KWallet
[2024-10-27 06:49:32,015] backend._load_plugins:205 DEBUG -> Loading SecretService
[2024-10-27 06:49:32,015] backend._load_plugins:205 DEBUG -> Loading Windows
[2024-10-27 06:49:32,031] __init__.<module>:11 DEBUG -> Loaded cffi backend
[2024-10-27 06:49:32,047] backend._load_plugins:205 DEBUG -> Loading chainer
[2024-10-27 06:49:32,047] backend._load_plugins:205 DEBUG -> Loading libsecret
[2024-10-27 06:49:32,047] backend._load_plugins:205 DEBUG -> Loading macOS
[2024-10-27 06:49:32,062] settings.__init__:14 DEBUG -> Settings filename: \HKEY_CURRENT_USER\Software\Buzz\OrganizationDefaults
[2024-10-27 06:49:32,062] settings.__init__:14 DEBUG -> Settings filename: \HKEY_CURRENT_USER\Software\Buzz\OrganizationDefaults
[2024-10-27 06:49:33,207] file_transcriber_queue_worker.run:56 DEBUG -> Starting next transcription task
[2024-10-27 06:49:33,366] file_transcriber.run:48 DEBUG -> Downloading audio file from URL: https://www.youtube.com/watch?v=30IpveNICwU
[2024-10-27 06:49:33,382] YoutubeDL.to_screen:939 DEBUG -> [youtube] Extracting URL: https://www.youtube.com/watch?v=30IpveNICwU
[2024-10-27 06:49:33,382] YoutubeDL.to_screen:939 DEBUG -> [youtube] 30IpveNICwU: Downloading webpage
[2024-10-27 06:49:34,528] YoutubeDL.to_screen:939 DEBUG -> [youtube] 30IpveNICwU: Downloading ios player API JSON
[2024-10-27 06:49:34,670] YoutubeDL.to_screen:939 DEBUG -> [youtube] 30IpveNICwU: Downloading web creator player API JSON
[2024-10-27 06:49:34,768] YoutubeDL.to_screen:939 DEBUG -> [youtube] 30IpveNICwU: Downloading m3u8 information
[2024-10-27 06:49:35,334] YoutubeDL.to_screen:939 DEBUG -> [info] 30IpveNICwU: Downloading 1 format(s): 251
[2024-10-27 06:49:35,431] YoutubeDL.to_screen:939 DEBUG -> [download] Destination: C:\Users\rosha\AppData\Local\Temp\tmpq5v215k4
[2024-10-27 06:49:35,431] minicurses.write:111 DEBUG -> [download]   0.0% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,431] minicurses.write:111 DEBUG -> [download]   0.0% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,431] minicurses.write:111 DEBUG -> [download]   0.1% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,431] minicurses.write:111 DEBUG -> [download]   0.1% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,431] minicurses.write:111 DEBUG -> [download]   0.3% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,445] minicurses.write:111 DEBUG -> [download]   0.6% of   10.50MiB at    4.34MiB/s ETA 00:02
[2024-10-27 06:49:35,445] minicurses.write:111 DEBUG -> [download]   1.2% of   10.50MiB at    8.75MiB/s ETA 00:01
[2024-10-27 06:49:35,463] minicurses.write:111 DEBUG -> [download]   2.4% of   10.50MiB at    7.62MiB/s ETA 00:01
[2024-10-27 06:49:35,463] minicurses.write:111 DEBUG -> [download]   4.8% of   10.50MiB at   15.27MiB/s ETA 00:00
[2024-10-27 06:49:35,476] minicurses.write:111 DEBUG -> [download]   9.5% of   10.50MiB at   21.96MiB/s ETA 00:00
[2024-10-27 06:49:35,492] minicurses.write:111 DEBUG -> [download]  19.0% of   10.50MiB at   32.70MiB/s ETA 00:00
[2024-10-27 06:49:35,523] minicurses.write:111 DEBUG -> [download]  38.1% of   10.50MiB at   43.15MiB/s ETA 00:00
[2024-10-27 06:49:35,618] minicurses.write:111 DEBUG -> [download]  76.2% of   10.50MiB at   42.69MiB/s ETA 00:00
[2024-10-27 06:49:35,652] minicurses.write:111 DEBUG -> [download]  92.7% of   10.50MiB at   43.88MiB/s ETA 00:00
[2024-10-27 06:49:35,666] minicurses.write:111 DEBUG -> [download]  92.7% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,666] minicurses.write:111 DEBUG -> [download]  92.7% of   10.50MiB at  Unknown B/s ETA Unknown
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  92.8% of   10.50MiB at  445.66KiB/s ETA 00:01
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  92.8% of   10.50MiB at  954.99KiB/s ETA 00:00
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  93.0% of   10.50MiB at    1.93MiB/s ETA 00:00
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  93.3% of   10.50MiB at    3.92MiB/s ETA 00:00
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  93.9% of   10.50MiB at    7.90MiB/s ETA 00:00
[2024-10-27 06:49:35,681] minicurses.write:111 DEBUG -> [download]  95.1% of   10.50MiB at   15.85MiB/s ETA 00:00
[2024-10-27 06:49:35,697] minicurses.write:111 DEBUG -> [download]  97.4% of   10.50MiB at   15.80MiB/s ETA 00:00
[2024-10-27 06:49:35,697] minicurses.write:111 DEBUG -> [download] 100.0% of   10.50MiB at   24.29MiB/s ETA 00:00
[2024-10-27 06:49:35,697] minicurses.write:111 DEBUG -> [download] 100% of   10.50MiB in 00:00:00 at 30.23MiB/s
[2024-10-27 06:49:36,387] file_transcriber.run:73 DEBUG -> Downloaded audio to file: C:\Users\rosha\AppData\Local\Temp\tmpq5v215k4.wav
[2024-10-27 06:49:36,387] whisper_file_transcriber.transcribe:52 DEBUG -> Starting whisper file transcription, task = FileTranscriptionTask(transcription_options=TranscriptionOptions(language='en', task=<Task.TRANSCRIBE: 'transcribe'>, model=TranscriptionModel(), word_level_timings=False, temperature=(0.0, 0.2, 0.4, 0.6, 0.8, 1.0), initial_prompt='', openai_access_token='', enable_llm_translation=False, llm_prompt='', llm_model=''), file_transcription_options=FileTranscriptionOptions(file_paths=None, url='https://www.youtube.com/watch?v=30IpveNICwU', output_formats=set()), model_path='C:\\Users\\rosha\\AppData\\Local\\Buzz\\Buzz\\Cache\\models\\whisper\\large-v3.pt', id=49737809, uid=UUID('d1405ed7-e658-407e-b6dc-0ce245e6124c'), segments=[], status=None, error=None, queued_at=None, started_at=None, completed_at=None, output_directory=None, source=<Source.URL_IMPORT: 'url_import'>, file_path='C:\\Users\\rosha\\AppData\\Local\\Temp\\tmpq5v215k4.wav', url='https://www.youtube.com/watch?v=30IpveNICwU', fraction_downloaded=0.0)
[2024-10-27 06:49:36,403] whisper_file_transcriber.transcribe:57 DEBUG -> CUDA version detected: 12.1
[2024-10-27 06:49:56,655] whisper_file_transcriber.transcribe:78 DEBUG -> whisper process completed with code = 0, time taken = 0:00:20.265683, number of segments = 12
[2024-10-27 06:49:56,655] file_transcriber_queue_worker.run:40 DEBUG -> Waiting for next transcription task
raivisdejus commented 4 weeks ago

@mriamnobody Thanks for the detailed bug report. This has been fixed and will be available in the next release. You can get the fixed version from our latest development builds.

  1. Go to actions section https://github.com/chidiwilliams/buzz/actions/workflows/ci.yml?query=branch%3Amain
  2. Select any of the latest builds
  3. Scroll down to the Artifacts section and download the installation file. You need to be logged in tot he Github to see the download link.
mriamnobody commented 4 weeks ago

@mriamnobody Thanks for the detailed bug report. This has been fixed and will be available in the next release. You can get the fixed version from our latest development builds.

  1. Go to actions section https://github.com/chidiwilliams/buzz/actions/workflows/ci.yml?query=branch%3Amain
  2. Select any of the latest builds
  3. Scroll down to the Artifacts section and download the installation file. You need to be logged in tot he Github to see the download link.

Much appreciated @raivisdejus. I'm sorry I forgot to mention but I'm on latest build (v1.2.0).

raivisdejus commented 4 weeks ago

@mriamnobody Indeed, there was a bug! The very latest development build should import and transcribe urls correctly.