Commit History

4 days ago

  1. Drop the comment on ssr_mode Running

    Verified signature
    6f3a894
  2. Shorten the comment on ssr_mode

    Verified signature
    3d8a890
  3. Launch without server-side rendering⁠

    Verified signature
    306b9ee

11 days ago

  1. Add domain tags for the Home tab filters of HARP

    Verified signature
    e619a16

14 days ago

  1. Return an AIFF input as WAV⁠

    harp-dev +1: Claude Opus 5.5
    Verified signature
    6e74851

15 days ago

  1. Refuse an empty file before the GPU is reserved⁠

    harp-dev +1: Claude Opus 5.5
    Verified signature
    4ceb571
  2. Move the model card ahead of the model loading⁠

    harp-dev +1: Claude Opus 5.5
    Verified signature
    c31ecc7
  3. Keep a lossless container⁠

    harp-dev +1: Claude Opus 5.5
    Verified signature
    4c6f562
  4. Drop the length cap on the audio, keeping the reservation sized to it⁠

    Verified signature
    2e68f3b

16 days ago

  1. Reserve GPU time by chunk, as measured, and declare the Python version the Space runs.

    Verified signature
    ed17409
  2. Move to ZeroGPU: the Gradio SDK held at 6.17.3, torch 2.8.0, the pinned checkpoint built once in the worker, and requests checked and sized before the GPU.

    Verified signature
    f626c4a

17 days ago

  1. Pin pyharp v0.4.2.

    Verified signature
    0b16eb7

19 days ago

  1. Label the output with what it holds, which HARP shows as its name.

    Verified signature
    ac5cb0b

23 days ago

  1. Pin pyharp v0.4.1 by commit, and remove a generator leftover nothing reads.

    Verified signature
    8dc8668

26 days ago

  1. Tidy the launch call and drop stray files

    Verified signature
    40b9ecb
  2. Roll pyharp and tidy the card description

    Verified signature
    f90a300
  3. Roll pyharp to v0.4.1

    Verified signature
    e2576dc
  4. Prune tags that predate the taxonomy

    Verified signature
    a6b25d0
  5. Copy the tag list from the running Space

    Verified signature
    592bf7c
  6. Declare the model's own sample rate and channel count

    Verified signature
    3ce8224

27 days ago

  1. Restore the input's sample rate and channel count

    Verified signature
    10ef29b
  2. Describe what each control does to the sound

    Verified signature
    f4061db
  3. Remove committed bytecode and editor metadata

    Verified signature
    8e96ecd
  4. Credit the original authors and link the repository and paper

    Verified signature
    e99434d
  5. Record the model's tags, as reported by /controls

    Verified signature
    c69d60f
  6. Fix the launch keywords and the taxonomy import

    Verified signature
    3d8ce40
  7. Bring the Space up to the pyharp example standard (taxonomy tags, __main__ guard, show_error, v0.4.0)

    Verified signature
    295b938
  8. Correct the license and give the Space an apt emoji and gradient

    Verified signature
    3deeab7

28 days ago

  1. Bring Space metadata up to HARP deployment guidelines

    Verified signature
    8f4b9e9

  1. Monkeypatch lowpass_filtering_prepare_inference boundary bug⁠

    997cf76

  1. Resample low-sample-rate input to 48kHz before inference⁠

    2e960ad

  1. add multiband ensemble control, clean up debug prints

    9a89e2a

  1. add torchcodec

    912d93c
  2. add server name

    c053253
  3. add startup debug prints

    8f7580c
  4. add startup debug prints

    55c1452
  5. add startup debug prints

    0378e6f
  6. add startup debug prints

    9c8e5e1
  7. add startup debug prints

    d28ce35
  8. lazy load model on first request

    53ce9f0

  1. add flush=True to print statements

    c3d0f7a
  2. add gitignore

    09d2e4b
  3. use super_resolution_long_audio from repo

    e89501c

  1. remove latent_t_per_second hardcode, use library default

    51984cc

  1. add guidance_scale and seed controls, fix author, fix ddim steps range

    e13f695

  1. document hardware requirements

    ecf9345
  2. pre-download model weights at build time

    0928555
  3. remove numpy pin

    53d85e1
  4. add git to dockerfile

    a440f6a
  5. switch to docker

    c5aff80