Posts

• Textgrid durations

  Get max/min segment durations from textgrids

  Aug 19, 2024

• Mapping (CMUdict) ARPAbet to Oculus visemes

  For a colleague

  Aug 10, 2024

• MT trained on crawled data still sucks

  Common Crawl contains a lot of Google Translate output. See if you can guess the source material

  Aug 2, 2024

• WhisperX using diarisation instead of VAD

  Monkey patched WhisperX with changed segmentation

  Jul 26, 2024

• Vosk CLI stderr output to CTM

  Because it was quicker than looking at the API examples

  Jul 25, 2024

• OWSM-CTC with CTCSegmentation for Irish

  tl;dr: OWSM-CTC is good enough for alignment for Irish

  Jun 27, 2024

• Generate seanchló data

  Creating synthetic data for training

  Jun 19, 2024

• Example of converting and using a Huggingface Whisper model with whisper.cpp

  For a student project

  Mar 3, 2024

• Recreating a phonetic dictionary with piper_phonemize

  For a student project

  Feb 29, 2024

• Split sentences in CTM-edit files

  Generating sentences from Riksdag: in progress

  Feb 26, 2024

• Load Sámi Whisper model

  Also basic pieces for scraping Sveriges Radio pages

  Feb 17, 2024

• Create n-gram LM from Dubliners

  For a student project

  Feb 16, 2024

• Dubliners - download audio and run ASR

  Runs ASR + phonetic recognition on two versions of Dubliners from Librivox: one (v2) with correct pronunciations, the other read by Americans

  Feb 15, 2024

• Waxholm lexicon processing

  I can't remember what this was for; I'm sure I'll be reminded

  Dec 15, 2023

• Waxholm kludge

  Reading old data

  Oct 17, 2023