11 stories tagged Speech Recognition
IBM has released Granite Speech 5.0, a collection of speech-language models that offer improved accuracy, speed, and multilingual capabilities. These models cover transcription, translation, and speaker attribution for enterprise use.
TII UAE released Falcon ASR, an open-source speech recognition system that rivals commercial tools in accuracy and supports multiple languages and dialects.
Hugging Face added Swahili to the Open ASR Leaderboard, the first Global South language on the benchmark. This expands access to voice technology for over 200 million Swahili speakers across East Africa.
Researchers introduced DonorRank, a learning-to-rank framework that predicts the best 'donor' languages for training zero-shot automatic speech recognition (ASR) models, evaluated on Indic and African language corpora.
Hugging Face introduced the FFASR Leaderboard to evaluate speech recognition models in real-world conditions. This new tool helps developers compare how well AI understands everyday speech, not just lab recordings.
NVIDIA's Nemotron 3.5 ASR model can now be fine-tuned for specific languages, accents, or domains. This makes it more accurate for specialized applications like medical or legal transcription.
Researchers have developed a compact AI model called FormalASR that converts spoken Chinese into polished, formal written text in one step. This could make transcribing meetings, lectures, and interviews much easier and faster.
Hugging Face is updating its Open ASR Leaderboard to prevent overfitting by adding a new metric. This change aims to ensure models perform well on real-world data, not just on benchmarks.
Researchers found that discrete speech units (DSUs) struggle to reliably encode suprasegmental information like lexical tone. This poses challenges for tasks where prosody is crucial, such as text-to-speech and multimodal dialogue systems.
Researchers introduce Contextual Earnings-22, a benchmark highlighting the gap between academic and real-world speech recognition performance. The study emphasizes the importance of contextual conditioning in high-stakes domains.
Researchers have developed a hybrid CNN-Transformer architecture for Arabic Speech Emotion Recognition (SER), addressing the scarcity of annotated datasets in Arabic. This model leverages convolutional layers and Transformer architecture to improve emotion detection accuracy in speech.