Research

Paper

AI LLM February 23, 2026

CTC-TTS: LLM-based dual-streaming text-to-speech with CTC alignment

Authors

Hanwen Liu, Saierdaer Yusuyin, Hao Huang, Zhijian Ou

Abstract

Large-language-model (LLM)-based text-to-speech (TTS) systems can generate natural speech, but most are not designed for low-latency dual-streaming synthesis. High-quality dual-streaming TTS depends on accurate text--speech alignment and well-designed training sequences that balance synthesis quality and latency. Prior work often relies on GMM-HMM based forced-alignment toolkits (e.g., MFA), which are pipeline-heavy and less flexible than neural aligners; fixed-ratio interleaving of text and speech tokens struggles to capture text--speech alignment regularities. We propose CTC-TTS, which replaces MFA with a CTC based aligner and introduces a bi-word based interleaving strategy. Two variants are designed: CTC-TTS-L (token concatenation along the sequence length) for higher quality and CTC-TTS-F (embedding stacking along the feature dimension) for lower latency. Experiments show that CTC-TTS outperforms fixed-ratio interleaving and MFA-based baselines on streaming synthesis and zero-shot tasks. Speech samples are available at https://ctctts.github.io/.

Metadata

arXiv ID: 2602.19574
Provider: ARXIV
Primary Category: eess.AS
Published: 2026-02-23
Fetched: 2026-02-24 04:38

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.19574v1</id>\n    <title>CTC-TTS: LLM-based dual-streaming text-to-speech with CTC alignment</title>\n    <updated>2026-02-23T07:44:14Z</updated>\n    <link href='https://arxiv.org/abs/2602.19574v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.19574v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Large-language-model (LLM)-based text-to-speech (TTS) systems can generate natural speech, but most are not designed for low-latency dual-streaming synthesis. High-quality dual-streaming TTS depends on accurate text--speech alignment and well-designed training sequences that balance synthesis quality and latency. Prior work often relies on GMM-HMM based forced-alignment toolkits (e.g., MFA), which are pipeline-heavy and less flexible than neural aligners; fixed-ratio interleaving of text and speech tokens struggles to capture text--speech alignment regularities. We propose CTC-TTS, which replaces MFA with a CTC based aligner and introduces a bi-word based interleaving strategy. Two variants are designed: CTC-TTS-L (token concatenation along the sequence length) for higher quality and CTC-TTS-F (embedding stacking along the feature dimension) for lower latency. Experiments show that CTC-TTS outperforms fixed-ratio interleaving and MFA-based baselines on streaming synthesis and zero-shot tasks. Speech samples are available at https://ctctts.github.io/.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.AS'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SD'/>\n    <published>2026-02-23T07:44:14Z</published>\n    <arxiv:comment>Submitted to INTERSPEECH 2026</arxiv:comment>\n    <arxiv:primary_category term='eess.AS'/>\n    <author>\n      <name>Hanwen Liu</name>\n    </author>\n    <author>\n      <name>Saierdaer Yusuyin</name>\n    </author>\n    <author>\n      <name>Hao Huang</name>\n    </author>\n    <author>\n      <name>Zhijian Ou</name>\n    </author>\n  </entry>"
}