SEAM: Spectral Embedding Alignment and Merging for Low-Resource TTS

Researchers introduced SEAM (Spectral Embedding Alignment and Merging), a zero-audio initialization method for expanding the vocabulary of conditional text-to-speech models without paired-audio fine-tuning. SEAM targets the bottleneck that extensive fine-tuning with paired audio creates for low-resource and cross-lingual TTS adaptation.

Expanding the vocabulary of conditional text-to-speech TTS models typically requires extensive fine-tuning with paired audio, creating a major bottleneck for low-resource and cross-lingual adaptation. To address this, we introduce SEAM, a zero-audio initialization method that constructs embeddings