Stack Overflow
Stack Overflow

Questions tagged [openai-whisper]

Whisper is a general-purpose speech recognition library by OpenAI.

175 questions
0
votes
0 answers

'text' field in segments sometimes has single and sometimes double quotes

The result of the transcribe function has a field called segments. Each segment contains a text accessible under the key text. For some segments, the quotes around the text are " ", for some it is ' '. Why is that the case, and what is the best way…
Moritz Groß
  • 1,352
  • 12
  • 30
0
votes
1 answer

Permission denied/ffmpeg error with Speech Recognition library

I'm trying to use the speech recognition library to set up a voice assistant, but I keep receiving a permission denied error. I suppose that the error is that the file isn't saved at any location where Python can access it, but I'm not entirely sure…
sally carlund
  • 1
  • 1
0
votes
0 answers

Google Colab - unable to use !pip - /bin/bash: pip install ***: No such file or directory

One of my colleagues is unable to install python packages within Google Colab. For example (run as a first statement in a clean session): !pip install git+https://github.com/openai/whisper.git This should install Open AI's Whisper. However it…
Richard Allen
  • 1
0
votes
0 answers

Huge output message

I'm trying to execute the following code using tensorflow, Hugginface's transformer and openai/whisper-base model import tensorflow as tf import transformers # Load the model and tokenizer model =…
Rendo Hugh Mann Sama
  • 1
0
votes
0 answers

Can't create an app or a script using Whisper from OpenAi

I have a simple Python code that I can't export as an app using py2app nor as an executable script with pyinstaller...I get "RecursionError: maximum recursion depth exceeded". Pyinstaller tells me to use the .spec file and to add "import sys ;…
HNT
  • 15
  • 3
0
votes
1 answer

How to connect to local runtime in Google Colab for this specific notebook WhisperWithVAD

I wish someone could help me to connect local runtime on this specific notebook on Google Colab at this link: https://colab.research.google.com/github/ANonEntity/WhisperWithVAD/blob/main/WhisperWithVAD.ipynb Basically it is a modified version of…
cendahoang
  • 1
0
votes
0 answers

How to set up a suctom dataset to train on whisper

How can I set tup a dataset to train on whisper model with custom data, I have transcripctions on excel file and audios in another folder, all the examples call a dataset from huggingFace
jhoanmartinez
  • 71
  • 7
0
votes
0 answers

Parallel inference on single model in CUDA cause worker processes to terminate

I'm trying to start openai\whisper inference on single model in CUDA with multiprocessing.Pool. On 6 workers inference works fine, except of some CUDA warnings on exiting worker processes. On 7 and more workers all works, but at the end (when some…
gorb
  • 1
  • 1
0
votes
0 answers

Whisper Open Ai Got Stuck

import whisper import os model = whisper.load_model("medium") print("Sarted") result = model.transcribe(os.path.join("samples","speech.mp3"),fp16=False) print(result["text"]) You can find your API key on https://replicate.com (Whisper_api)…
Prasun Kumar
  • 11
  • 3
0
votes
0 answers

Cannot find specified path: Whisper Error, python 3.7.7

Traceback (most recent call last): File "C:\Users\SAMSUNG\Desktop\Whisper\test.py", line 4, in audio = whisper.load_audio("audio.mp3") File "C:\Users\SAMSUNG\AppData\Local\Programs\Python\Python37\lib\site-packages\whisper\audio.py", line 44, in…
Ammar Ahmed Siddiqui
  • 23
  • 1
  • 5
0
votes
0 answers

pyqt5 gui dependent on ffmpeg compiled with pyinstaller doesn't run on other machines?

I am trying to create a simple Pyqt5 GUI for Windows 10 that uses OpenAI's model Whisper to transcribe a sound file and outputting the results in an Excel-file. It works on my own computer where I have installed the necessary dependencies for…
Soren
  • 17
  • 3
0
votes
2 answers

Converting Python Speech Recognition Audio Frame data to a numpy array that can be processed by Whisper?

I'm using the speech recognition Python library to record audio bytes from my microphone in mono at 16khz but I want to use the new Whisper library that accepts NumPy arrays, spectrograms, and file paths. Writing to a file takes too long so I'd like…
potatohprogrammer
  • 1
  • 1
0
votes
1 answer

File not found by Python openai/whisper

I've tried various arrangements of this file path name and an ls shows that the file is there. For simplicity, I'm in the directory as the file itself at the moment. I've stuck the print statement in client.py because that's where it's…
Hugh Barnard
  • 352
  • 2
  • 12
-1
votes
0 answers

Detect if whisper compatible GPU

In a .NET 6.0 project I am using this C# port to transcribe audio to text. On a PC with a GeForce 8600, the application crashes completely when calling Library.loadModel(). From what I have found, Whisper requires 1GB of video memory and D3D…
stefan
  • 195
  • 1
  • 13
-1
votes
0 answers

Whisper Large-V2 model taking Majority of VRAM

I was testing out the Large-V2 Whisper Model by OpenAI and was wanted to test it out on a GPU (Specifically a MSI GTX 960 4GBVRAM). After I run the following code: model_size = "large-v2" # CPU with INT8 model = WhisperModel(model_size,…
abdz_128
  • 31
  • 8
1 2 3
…
11
12