{
  "id": 4981658,
  "title": "Why AI Models Sometimes Elongate Their Greetings Like 'Kon'nichiwa~'",
  "url": "https://urgent.news/2026/09/02/why-ai-models-sometimes-elongate-their-greetings-like-konnichiwa",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-02T00:53:08.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/orca_forge/why-ai-models-sometimes-elongate-their-greetings-like-konnichiwa-ole"
  },
  "original_language": "en",
  "account": "When the trained voice model pronounced the Japanese greeting こんにちは, it elongated the phrase to こんにちわー. The transcription was performed using Whisper, which transcribed the audio and compared it with a script, but there was no instruction in the script to stretch the greeting. The feedback indicated that the transcription read こんにちわー with an accent on the last syllable, suggesting something was mixed in.\n\nThe training corpus contained clips with stretched endings, but the mechanism to detect this was fundamentally non-functional by design. Script matching was done by converting both the script and transcription to kana (hiragana) and comparing them. However, the normalization process removed prolongation marks (the ー character), making it impossible to detect stretched endings even if they were present in the raw transcription from Whisper.\n\nThe normalization was intended to absorb variations in notation and check for content consistency, but it also discarded information about stretched endings. The same issue occurred with consecutive vowels. The problem arose when trying to separate the judgment of content consistency from the judgment of stretched endings, as the raw transcription needed to be passed to the latter function for accurate detection.",
  "summary": "📝 Originally published (in Japanese) at forge.workstyle.tech . When I had the trained voice model read \"こんにちは\" (Hello), it stretched the phrase to \"こんにちわぁ.\" There was no instruction to stretch it in the script. The feedback was as follows: For \"こんにちは,\" it's pronounced as \"こんにちわぁ\" with an accent on the last syllable. It feels like something is mixed in. \"Something mixed in\" was accurate, and…",
  "key_points": [
    "Whisper transcription elongated こんにちは to こんにちわー",
    "No instruction in script to stretch greeting",
    "Stretched endings undetectable due to normalization process"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}