Skip to content

Conversation

@mkopcins
Copy link
Collaborator

Description

The Whisper model export now takes in a plain waveform instead of pre-computed STFT. This PR aims to change the current API to accept waveforms instead. Before merging this, make sure to re-export all the existing Whisper models with the new export script.

Introduces a breaking change?

  • Yes
  • No

Type of change

  • Bug fix (change which fixes an issue)
  • New feature (change which adds functionality)
  • Documentation update (improves or adds clarity to existing documentation)
  • Other (chores, tests, code style improvements etc.)

Tested on

  • iOS
  • Android

Testing instructions

Screenshots

Related issues

Checklist

  • I have performed a self-review of my code
  • I have commented my code, particularly in hard-to-understand areas
  • I have updated the documentation accordingly
  • My changes generate no new warnings

Additional notes

constexpr int32_t innerDim = 256;

std::vector<float> preprocessedData =
dsp::stftFromWaveform(waveform, fftWindowSize, stftHopLength);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If stftFromWaveform will be removed, this function should be removed completely from codebase since it is not used anywhere else.

@msluszniak msluszniak added the feature PRs that implement a new feature label Oct 23, 2025
@msluszniak msluszniak linked an issue Oct 23, 2025 that may be closed by this pull request
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

feature PRs that implement a new feature

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Remove STFT calculation in STT

4 participants