Комментарии:
Any updates to this?
Ответитьis it possibe to have label like speaker a or speaker b.. to label who said this or that
ОтветитьIs there any windows app based on this for offline transcription using audio (mp3, wav) files as inputs?
Ответитьcan you please share the python code link
ОтветитьThank you so much!
ОтветитьLate comment, but Whisper even surpasses VOSK api (which was the best offline ASR engine prior to Whisper) which I used to use in some Python scripts for transcriptions and forced alignment.
ОтветитьBut i wonder y would one need in Text when its easier to listen audio.. :)
ОтветитьI've used this for Spanish language videos and on some of the transcript it repeats words even though the next dialogue is completely different.
ОтветитьHow do is convert these transcription to srt file anyone help please
ОтветитьIs there any way to use this Whisper-jax in python. I am trying to have a program that transcripts audio files for me and would like to use this to do so. Is it possible to do so.
ОтветитьHey man, just subscribed. Normally I see a bunch of hype or garbage ass videos, but yours just gives a nice quick overview, no bs on the latest AI stuff. So I subscribed. Thanks.
ОтветитьThis is brilliant 🤩 I am curious if we wanted to isolate and identify each individual speaker in an audio, how we go about doing that? What kind of libraries would we use?
For example: If the audio has 3 individuals, I want to identify and label each individual as Person 1-3.
Ideally, if the person introduces themselves or is introduced in the audio label the part of the audio to that person. Adobe Audition, Premiere do something similar.
Can we perhaps us Meta's Segment Anything or is that only for images?
Would love to hear your thoughts
HUGGING FACE IS NOLONGER WORKING ITS SAYING TIMEOUT ERROR 502
ОтветитьHi sir, thanks a lot for this tutorial, can you show me how to input my own audio file?
ОтветитьDoes this work with any mp3 file and does the program save the files onto a seperate server or are these files deleted after a few hours?
Ответитьare there any gains of using jax on cpu compared to original model whisper on cpu?
ОтветитьCan anyone help me setting this up? And I’ll pay them of course
ОтветитьHow can I modify the code to upload my own .mp3 file? It appears pretty straightforward according to the instructions of the Kaggle notebook: "Note that you can also pass any .mp3, .wav, or .flac audio file directly to the Whisper JAX pipeline, and it will automatically load the audio file for you." However, I don't know how to code. Please help.
ОтветитьThis is awesome. Any guidance on transcribing FB live videos? Do i need to download them first and then have whisper transcribe them?
ОтветитьCan you run it on Pixel with tensor TPU?
ОтветитьIs it possible to run (with GUI) on your own computer like GITHUB Const-me
Whisper ?
your awesome
ОтветитьWhen i use whisper Jax on Huggingface what model i am using? i normally like to use the large one
Ответитьhuggingface link: 502 Bad Gateway
Ответитьhugging face link not working 502 bad gateway
ОтветитьAny GPU version available which can transcribe faster than openai versiob
Ответитьis there any model available who can transcript Bangla Language?
ОтветитьThis is spectacular! Thanks for a great video. I was looking for something like this.
ОтветитьThis is going to be revolutionary in medical transcription.
ОтветитьIs it possible to use this for live transcription?.
Ответитьis there a cost to use this on my own application?
ОтветитьDo you know if this is runnable locally?
ОтветитьWhat TPU version does It support?
ОтветитьWill it transcript voice to text on multiple languages?
Ответитьamazing
ОтветитьThanks.
Ответить