Small-e: Small Language Model With Linear Attention For Efficient Speech Synthesis

Abstract

Recent advancements in text-to-speech (TTS) powered by language models have showcased remarkable capabilities in achieving naturalness and zero-shot voice cloning. Notably, the decoder-only transformer is the prominent architecture in this domain. However, transformers face challenges stemming from their quadratic complexity in sequence length, impeding training on lengthy sequences and resource-constrained hardware. Moreover they lack specific inductive bias with regards to the monotonic nature of TTS alignments. In response, we propose to replace transformers with emerging recurrent architectures and introduce specialized cross-attention mechanisms for reducing repeating and skipping issues. Consequently our architecture can be efficiently trained on long samples and achieve state-of-the-art zero-shot voice cloning against baselines of comparable size. Our implementation and demos are available at https://github.com/theodorblackbird/lina-speech.

Stats

• citations3
• S2 citations—
• github stars178
• HF likes0
• heat score9.02
• arxiv keylemerle2024small

Code

• theodorblackbird/lina-speech—

Related papers

• Lina-speech: Gated Linear Attention And Initial-state Tuning For Multi-sample Prompting Text-to-speech Synthesis (2024)4.51
• Improving Robustness Of Llm-based Speech Synthesis By Learning Monotonic Alignment (2024)0.00
• Attentron: Few-shot Text-to-speech Utilizing Attention-based Variable-length Embedding (2020)12.02
• Clam-tts: Improving Neural Codec Language Model For Zero-shot Text-to-speech (2024)0.00
• Tacolm: Gated Attention Equipped Codec Language Model Are Efficient Zero-shot Text To Speech Synthesizers (2024)0.00
• Towards Lifelong Learning Of Multilingual Text-to-speech Synthesis (2021)3.58
• HALL-E: Hierarchical Neural Codec Language Model For Minute-long Zero-shot Text-to-speech Synthesis (2024)0.00
• Self-attention Linguistic-acoustic Decoder (2018)2.26