5 papers
RVQ Position Aware Speculative Decoding for On Device Text to Speech
Berkin Durmus, Eduardo Pacheco, Zach Nagengast +1
Autoregressive decoding (AR) with Transformer models is memory bandwidth bound at single stream inference, the typical deployment regime for on device text to speech (TTS). Real ti…
Taming Long-form Text-to-Speech
Rongxiang Wang, Berkin Durmus, Aysegul Orhon +2
Long-form text-to-speech (TTS) enables multi-turn conversations with consistent prosody and higher quality voice cloning from longer reference audio. Recent open-weights autoregres…
Contextual Earnings-22: A Speech Recognition Benchmark with Custom Vocabulary in the Wild
Berkin Durmus, Chen Cen, Eduardo Pacheco +2
The accuracy frontier of speech-to-text systems has plateaued on academic benchmarks.1 In contrast, industrial benchmarks and adoption in high-stakes domains suggest otherwise. We…
SDBench: A Comprehensive Benchmark Suite for Speaker Diarization
Eduardo Pacheco, Atila Orhon, Berkin Durmus +2
Even state-of-the-art speaker diarization systems exhibit high variance in error rates across different datasets, representing numerous use cases and domains. Furthermore, comparin…
WhisperKit: On-device Real-time ASR with Billion-Scale Transformers
Atila Orhon, Arda Okan, Berkin Durmus +2
Real-time Automatic Speech Recognition (ASR) is a fundamental building block for many commercial applications of ML, including live captioning, dictation, meeting transcriptions, a…