Enhancing Naturalness and Expressiveness of Text-to-Speech Generated Speech through Prosodic Parameter Manipulation
This project aims to improve the naturalness and expressiveness of Text-to-Speech (TTS) systems by developing a machine learning model that manipulates the prosodic parameters (pitch, duration, and energy) of TTS-generated speech to make it more closely resemble human speech.
Wave-U-Mamba: An Efficient and High-Quality End-to-End Framework for Speech Super-Resolution
Wave-U-Mamba is an efficient and effective end-to-end framework for speech super-resolution that directly generates high-resolution speech waveforms from low-resolution inputs, outperforming existing state-of-the-art models.
Continuous Speech Separation and Transcription-Supported Diarization for Efficient Meeting Recognition
A modular pipeline for single-channel meeting transcription that combines continuous speech separation, automatic speech recognition, and transcription-supported diarization to achieve state-of-the-art performance.
A Lightweight Dual-Stage Framework for Personalized Speech Enhancement Based on DeepFilterNet2
A novel method to personalize a lightweight dual-stage speech enhancement model, DeepFilterNet2, by integrating speaker embeddings into the model architecture, achieving significant performance improvements while preserving minimal computational overhead.
Efficient Speech Processing with Discrete Speech Units: Techniques and Insights from the Interspeech 2024 Challenge
The authors present their systems developed for the Interspeech 2024 Speech Processing Using Discrete Speech Unit Challenge, including techniques for text-to-speech, singing voice synthesis, and automatic speech recognition using discrete speech tokens. Their approaches demonstrate the potential of discrete speech representations to achieve high-quality and low-bitrate speech processing.
Improving Automatic Speech Recognition in Noisy Environments: A Case Study on Cocktail Party Speech Recognition
This paper presents an end-to-end model that combines a speech enhancement module (ConVoiFilter) and an automatic speech recognition (ASR) module to improve speech recognition performance in noisy, crowded environments. The model utilizes a single-channel speech enhancement approach to isolate the target speaker's voice from background noise and then feeds the enhanced audio into the ASR module.
Leveraging Text-to-Speech Knowledge for Robust Open Vocabulary Keyword Spotting
A novel framework that leverages intermediate representations extracted from a pre-trained text-to-speech (TTS) model to enhance the performance of open vocabulary keyword spotting.
Robust Target-Speaker Voice Activity Detection Tolerant to Speaker Profile Errors
The proposed Profile-Error-Tolerant Target-Speaker Voice Activity Detection (PET-TSVAD) model is robust to speaker profile errors introduced in the first pass diarization, outperforming the existing TS-VAD models on both the VoxConverse and DIHARD-I datasets.
Efficient Neural Codec Language Modeling for Zero-Shot Text-to-Speech Synthesis
CLaM-TTS employs probabilistic residual vector quantization to achieve superior compression in token length and enable a language model to generate multiple tokens at once, thereby enhancing the efficiency of zero-shot text-to-speech synthesis.
Leveraging Whisper's Robust Speech Recognition for Predicting Microscopic Intelligibility of Noisy Speech
Transfer learning from the state-of-the-art Whisper automatic speech recognition model can effectively predict the distribution of lexical responses perceived by human listeners to noisy speech stimuli, outperforming baseline methods.
