A new multi-accent long-form call-center dialogue dataset for English ASR evaluation shows substantial performance variation across accents and segmentation methods.
Granite-speech: open-source speech-aware llms with strong english asr capabilities
12 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
12
Pith papers citing it
1
external citations · external index
representative citing papers
LLM decoders in speech recognition show no racial bias amplification and fewer repetition hallucinations under degradation than Whisper, with audio encoder design mattering more than model scale for fairness and robustness.
GLOW integrates a pre-trained GNN for candidate prediction with an LLM for joint symbolic-semantic reasoning over incomplete KGs, reporting up to 53.3% gains on standard benchmarks and a new GLOW-BENCH dataset.
Common-word acoustic cues and bias-word position prediction in speech LLMs cut rare-word transcription errors by 16.3% versus baselines, including out-of-domain cases.
With matched data and sizes, CTC+LLM shallow fusion beats tight speech-LLM integration on in-domain ASR, while prefix LLMs win average WER on out-of-domain HuggingFace sets.
Internal decoder probing of Whisper yields strongest hallucination detection without references, with late fusion of text and internal features performing best overall.
SALSA adapts speech-aware LLMs via supervised layer-wise steering vectors, reporting up to 46.8% relative gains over zero-shot on out-of-domain speech benchmarks.
A novel alignment algorithm using dynamic programming and beam search provides more accurate matching of individual errors between reference and model transcripts for improved speech recognition evaluation.
JSTIP interleaves speech and text sequences during pretraining on 38k hours of ASR data to improve entity accuracy over ASR-only and simple joint-training baselines while matching performance from domain text.
ESPnet3 introduces a new modular architecture with DataOrganizer and sharding to cut training time and simplify model integration for speech research.
Audio language models are benchmarked on five semantic and paralinguistic reasoning tasks to reveal limitations in handling spoken audio evidence, accent variation, and domain shifts.
Lightweight training strategies allow speech-aware LLMs to output accurate word timestamps alongside ASR transcripts while also improving recognition quality across datasets.
citing papers explorer
-
AppTek Call-Center Dialogues: A Multi-Accent Long-Form Benchmark for English ASR
A new multi-accent long-form call-center dialogue dataset for English ASR evaluation shows substantial performance variation across accents and segmentation methods.
-
Do LLM Decoders Listen Fairly? Benchmarking How Language Model Priors Shape Bias in Speech Recognition
LLM decoders in speech recognition show no racial bias amplification and fewer repetition hallucinations under degradation than Whisper, with audio encoder design mattering more than model scale for fairness and robustness.
-
Leveraging LLM-GNN Integration for Open-World Question Answering over Knowledge Graphs
GLOW integrates a pre-trained GNN for candidate prediction with an LLM for joint symbolic-semantic reasoning over incomplete KGs, reporting up to 53.3% gains on standard benchmarks and a new GLOW-BENCH dataset.
-
Contextual Biasing for ASR in Speech LLM with Common Word Cues and Bias Word Position Prediction
Common-word acoustic cues and bias-word position prediction in speech LLMs cut rare-word transcription errors by 16.3% versus baselines, including out-of-domain cases.
-
LLMs and Speech: Integration vs. Combination
With matched data and sizes, CTC+LLM shallow fusion beats tight speech-LLM integration on in-domain ASR, while prefix LLMs win average WER on out-of-domain HuggingFace sets.
-
From Text Metrics to Model Internals: A Study of Whisper ASR Hallucination Detection
Internal decoder probing of Whisper yields strongest hallucination detection without references, with late fusion of text and internal features performing best overall.
-
SALSA: Speech Aware LLM Adaptation via Learned Steering Activation Vectors
SALSA adapts speech-aware LLMs via supervised layer-wise steering vectors, reporting up to 46.8% relative gains over zero-shot on out-of-domain speech benchmarks.
-
A Text-To-Text Alignment Algorithm for Better Evaluation of Modern Speech Recognition Systems
A novel alignment algorithm using dynamic programming and beam search provides more accurate matching of individual errors between reference and model transcripts for improved speech recognition evaluation.
-
Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving
JSTIP interleaves speech and text sequences during pretraining on 38k hours of ASR data to improve entity accuracy over ASR-only and simple joint-training baselines while matching performance from domain text.
-
ESPnet3: Infrastructure for Scalable Speech and Audio Research in the Foundation Model Era
ESPnet3 introduces a new modular architecture with DataOrganizer and sharding to cut training time and simplify model integration for speech research.
-
Afrispeech Semantics: Evaluating Audio Semantic Reasoning in Spoken Language Models Across Domains and Accents
Audio language models are benchmarked on five semantic and paralinguistic reasoning tasks to reveal limitations in handling spoken audio evidence, accent variation, and domain shifts.
-
In-Sync: Adaptation of Speech Aware Large Language Models for ASR with Word Level Timestamp Predictions
Lightweight training strategies allow speech-aware LLMs to output accurate word timestamps alongside ASR transcripts while also improving recognition quality across datasets.