What do the VibeVoice authors say about long-form multi-speaker generation: length, speaker count, evaluation, limitations and misuse risks?

Problem: choose a model for podcast or multi-participant audio. Conditions: dialogue transcripts with speaker labels. Method (VibeVoice technical report): a 7.5 Hz continuous speech tokenizer plus an LLM with next-token diffusion; up to 90 minutes in a 64K context window with a maximum of four speakers. Evaluation (authors'): 8 long transcripts totalling about 1 hour; WER via Whisper-large-v3 and Nemo, speaker similarity via WavLM, and MOS from 24 annotators on realism, richness and preference, each listening to about six hours; comparisons include ElevenLabs v3 alpha and Gemini 2.5 Pro TTS. Stated limits: English and Chinese only; no non-speech audio (background noise, music, sound effects); no overlapping speech; deepfake and disinformation risk; the authors do not recommend commercial or real-world use without further testing and describe the model as research and development only. Failure modes: unexpected audio for other languages; long output drift is not separately measured in the extracted text. Limitations: prior item C2-B68-V4-07 covers consent/watermark enforcement; the compact test set means results are indicative. Version/date: arXiv 2508.19205. Curator qualifications: All cited documentation excerpts located on the live pages by curator GET; no video sources. Limitations: Evaluation numbers are the authors' own; the report text as extracted spells the model name with a space. Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.

{
  "citations": [
    {
      "url": "https://arxiv.org/pdf/2508.19205",
      "note": "Abstract; Evaluation; Limitations; Limitations; Limitations"
    }
  ]
}