speech recognition
PulseAugur coverage of speech recognition — every cluster mentioning speech recognition across labs, papers, and developer communities, ranked by signal.
- instance of alphaXiv 90%
- instance of CatalyzeX 90%
- developed Speech LLMs 90%
- affiliated with speech synthesis 70%
- instance of ScienceCast 70%
- instance of Gotit.pub 70%
- used by CatalyzeX 70%
- used by word error rate 70%
- used by LibriSpeech 70%
- instance of Speech LLMs 70%
- instance of AI Speech Technologies 70%
- used by whisper tiny 70%
18 day(s) with sentiment data
-
LokalBot integrates five local AI models for Mac meeting productivity
LokalBot is an open-source Mac application designed to orchestrate multiple local AI models for a variety of tasks. The app can record meetings, transcribe them, generate notes, enable searching, and provide autocomplet…
-
KlangAI releases state-of-the-art Danish ASR model pianissimo-da
KlangAI has released pianissimo-da, a new Danish Automatic Speech Recognition (ASR) model. This model is available in multiple formats including GGUF, ONNX, MLX, and CoreML, and is hosted on Hugging Face. It is describe…
-
AWS enables low-latency voice AI with State Space Models
AWS is enabling the development of ultra-low-latency voice AI applications through its infrastructure, partnering with companies like Cartesia. Cartesia specializes in real-time voice AI, utilizing State Space Models (S…
-
Musical intervention boosts infant neural processing of music and speech
A study suggests that musical interventions can improve how infants' brains process the temporal structures found in both music and speech. This research explores the potential benefits of early music exposure for cogni…
-
GhostPrefix attack manipulates ASR systems without synchronization
Researchers have developed GhostPrefix, a novel method for manipulating Automatic Speech Recognition (ASR) systems without requiring synchronization with the target audio. This technique involves prepending a short, reu…
-
New ASR research explores model merging and language adaptation techniques
Researchers are exploring advanced techniques to improve Automatic Speech Recognition (ASR) systems, particularly for low-resource languages and specialized domains. One study focuses on text-only domain adaptation for …
-
New GSR-Bench benchmark redefines speech recognition beyond transcription
Researchers have introduced Grounded Generative Speech Recognition (GSR), a new framework that moves beyond traditional transcription to evaluate speech-to-text systems. GSR reformulates speech recognition as generating…
-
AI Speech Technologies: TTS, STT, Voice Cloning, and Synthesis Notes
This Mastodon post serves as a curated collection of notes and links pertaining to AI speech technologies. It encompasses various facets of the field, such as Text-to-Speech (TTS), Speech-to-Text (STT), voice synthesis,…
-
Open-weight AI models surge in popularity on Hugging Face · 7 sources tracked
Several open-weight AI models are gaining significant traction on Hugging Face, as indicated by download and like counts over the past 30 days. Notable performers include DeepSeek-V4.1-Flash-1017 with over 170,000 downl…
-
Voice AI accent bias: Geordie vs. SSBE perceptions studied
A new study published on arXiv explores how users perceive and interact with voice AI systems that use regional accents. Researchers found that while a voice using Newcastle English (Geordie) was perceived as familiar a…
-
AI and Software Development Discussions Span Global Studios, Speech Recognition, and Privacy
This cluster aggregates several Mastodon posts discussing various aspects of AI and software development. One post details the founding of BalochDev, a remote-first software and AI studio. Another delves into the nuance…
-
New agentic framework optimizes multilingual ASR data sampling
Researchers have developed a new framework for multilingual Automatic Speech Recognition (ASR) that uses a language-aware agent to optimize data sampling. This agent analyzes ASR errors from linguistic perspectives and …
-
New GRPO method improves semi-supervised ASR accuracy
Researchers have proposed a new method called Group-Relative Policy Optimization (GRPO) for semi-supervised Automatic Speech Recognition (ASR). This approach uses pseudo-labels generated by a teacher model as a reward r…
-
New Text Injection Method Speeds Speech Recognition Training
Researchers have developed a new method for training speech recognition models using unpaired text, which speeds up training by 1.8 times compared to previous approaches. This technique, called "Text Injection," replace…
-
Mastodon user curates AI resources on speech, images, agents, and general AI
The user "taoofmac" on Mastodon has curated a series of notes and links covering various aspects of artificial intelligence. These collections include resources on AI speech technologies like Text-to-Speech and Speech-t…
-
New routing method enhances ASR domain adaptation without performance loss
Researchers have developed a new method called conservative residual routing to improve domain adaptation for Automatic Speech Recognition (ASR) systems. This technique allows an adapted predictor to act as a residual c…
-
Study reveals ASR bias against Newcastle English dialect
A new study published on arXiv analyzes bias in Automatic Speech Recognition (ASR) systems, focusing on Newcastle English. Researchers found that state-of-the-art ASR systems exhibit significant transcription errors whe…
-
Research quantifies ASR errors in TTS evaluation
A new research paper explores the accuracy of text-to-speech (TTS) systems by analyzing the errors introduced by automatic speech recognition (ASR) systems used in their evaluation. The study found that ASR systems miss…
-
New research compares search complexity in end-to-end ASR models
A new research paper systematically compares the search space complexity and efficiency of end-to-end Automatic Speech Recognition (ASR) models, specifically Connectionist Temporal Classification (CTC) and neural transd…
-
New Rowhammer attack targets federated learning servers
Researchers have demonstrated a novel remote Rowhammer attack that targets the central server in federated learning systems. By manipulating clients to generate repetitive memory updates on the server, an adversarial ag…