Skip to content

Latest commit

 

History

History
41 lines (27 loc) · 1.21 KB

README.md

File metadata and controls

41 lines (27 loc) · 1.21 KB
Raw

Whisper API Streaming

OpenAI has recently published their Whisper model API for audio transcription.

Unfortunately, this API does not provide streaming capabilities. This project aims to provide a streaming interface to the OpenAI API.

Functionality

Currently only streaming of response is supported. If you need also streaming of input aduio, please open an issue and describe what you need, it should be easy to implement.

Installation

pip install git+https://github.com/gkorepanov/whisper-stream.git

CLI usage

To transcribe a file, run the following command:

OPENAI_API_KEY=<KEY> whisperstream /path/to/your/audio/file.ogg -l en

You can omit the language parameter, it will be detected automatically.

Usage

from whisperstream import atranscribe_streaming_simple
path = '/path/to/your/audio/file.ogg'
language, gen = await atranscribe_streaming_simple(path)

# language is a Lang object from [iso369 lib](https://github.com/LBeaudoux/iso639)
print(language.name)

async for segment in gen:
    # segment is an OpenAI Python API object
    # it has `start`, `end`, `text` attributes
    print(segment.text, end="")