Skip to content
This repository was archived by the owner on Jan 7, 2025. It is now read-only.
This repository was archived by the owner on Jan 7, 2025. It is now read-only.

Batch v1 vs Batch v2 sending raw audio data #88

Open

Description

@dzlandis

Hi there!

I noticed that in Batch v1 it is possible to send raw audio bytes for transcription, which is preferable for my use case. However, I noticed that in Batch v2, it appears that URI's are used, which based on my understanding, means that sending raw audio bytes is not supported in Batch v2. Is being able to send raw audio bytes via Batch v2 planned in the future? Or is it currently possible and I am missing something? I was hoping just to get some clarification regarding this as I want to use the newest API version, but not if it means a critical use case for me is missing.

Activity

  1. langma commented on Aug 10, 2023

    @langma
    Collaborator

    Hi! Thanks for asking. You are correct, the v1 API does support raw bytes input, and that feature is missing from v2. The immediate reason is that we wanted the v2.ProcessAudioSource to be more specific and thus understandable, whereas the v1.ProcessAudio handled all use cases in the same method.
    We do have the raw byte stream input support on the backlog for the v2 also, but the release schedule for that is not clear at the moment. We will keep v1 alive until the streaming support exists in v2 also, so you can continue using the v1 raw byte input for now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions