# Machine Speech Chain with One-shot Speaker Adaptation

## TL;DR

This paper introduces a speech chain model that incorporates one-shot speaker adaptation, enabling the system to recognize and synthesize speech for unseen speakers by learning from minimal speaker data, thus improving multi-speaker ASR and TTS performance.

## Contribution

The paper presents a novel speech chain framework that integrates speaker recognition and one-shot speaker adaptation, allowing the model to handle unseen speakers effectively.

## Key findings

- Enhanced speech recognition accuracy for unseen speakers.
- Successful voice synthesis mimicking new speakers with only one sample.
- Improved mutual learning between ASR and TTS components.

## Abstract

In previous work, we developed a closed-loop speech chain model based on deep learning, in which the architecture enabled the automatic speech recognition (ASR) and text-to-speech synthesis (TTS) components to mutually improve their performance. This was accomplished by the two parts teaching each other using both labeled and unlabeled data. This approach could significantly improve model performance within a single-speaker speech dataset, but only a slight increase could be gained in multi-speaker tasks. Furthermore, the model is still unable to handle unseen speakers. In this paper, we present a new speech chain mechanism by integrating a speaker recognition model inside the loop. We also propose extending the capability of TTS to handle unseen speakers by implementing one-shot speaker adaptation. This enables TTS to mimic voice characteristics from one speaker to another with only a one-shot speaker sample, even from a text without any speaker information. In the speech chain loop mechanism, ASR also benefits from the ability to further learn an arbitrary speaker's characteristics from the generated speech waveform, resulting in a significant improvement in the recognition rate.