How to set language with OpenAIWhisperAudio?

[enigmablue]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import { OpenAIWhisperAudio } from "langchain/document_loaders/fs/openai_whisper_audio";

const filePath = "./src/document_loaders/example_data/test.mp3";

const loader = new OpenAIWhisperAudio(filePath);

const docs = await loader.load();

console.log(docs);

Description

How do i set language to the OpenAIWhisperAudio?
Unfortunately even though i am speaking in english its giving me back transcriptions in the wrong language.

System Info

[email protected]
platform windows
node 10.5.0

[dosubot[bot]]

To set the language for transcriptions using the OpenAIWhisperAudio class from the langchain library, it appears there isn't a direct way to specify the language in the current implementation. The constructor for OpenAIWhisperAudio does not include a parameter for setting the language, and there's no global configuration option in langchain for setting a default language for operations like transcription.

This limitation suggests that the language detection and transcription process is managed internally by the OpenAI Whisper system, without explicit control over the language through the langchain library's API. If the transcriptions are not in the expected language, it might be beneficial to ensure the audio quality is clear, as this can significantly impact the accuracy of language detection and transcription.

Unfortunately, without a direct way to set the language in the langchain library, these are the options available. If clarity and audio quality adjustments do not resolve the issue, it may be necessary to look for alternative solutions or workarounds outside of the langchain library for language-specific transcription needs.

Sources

• examples/src/document_loaders/openai_whisper_audio.ts
• langchain/src/document_loaders/fs/openai_whisper_audio.ts

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.