Scribe
ElevenLabs has introduced Scribe, an advanced Automatic Speech Recognition (ASR) model designed to deliver highly accurate transcriptions across 99 languages. Scribe is engineered to handle diverse real-world audio scenarios, providing features such as word-level timestamps, speaker diarization, and audio-event tagging. Benchmark tests, including FLEURS and Common Voice, demonstrate Scribe's superior performance over leading models like Gemini 2.0 Flash, Whisper Large V3, and Deepgram Nova-3, achieving the lowest word error rates in languages such as Italian (98.7%) and English (96.7%). Notably, Scribe also significantly reduces errors in languages that have been traditionally underserved, including Serbian, Cantonese, and Malayalam, where other models often exhibit error rates exceeding 40%. Developers can integrate Scribe through ElevenLabs' speech-to-text API, receiving structured JSON transcripts that include detailed annotations.
Learn more
Amazon Nova
Amazon Nova is a new generation of state-of-the-art (SOTA) foundation models (FMs) that deliver frontier intelligence and industry leading price-performance, available exclusively on Amazon Bedrock.
Amazon Nova Micro, Amazon Nova Lite, and Amazon Nova Pro are understanding models that accept text, image, or video inputs and generate text output. They provide a broad selection of capability, accuracy, speed, and cost operation points.
Amazon Nova Micro is a text only model that delivers the lowest latency responses at very low cost.
Amazon Nova Lite is a very low-cost multimodal model that is lightning fast for processing image, video, and text inputs.
Amazon Nova Pro is a highly capable multimodal model with the best combination of accuracy, speed, and cost for a wide range of tasks. Amazon Nova Pro’s capabilities, coupled with its industry-leading speed and cost efficiency, makes it a compelling model for almost any task, including video summarization, Q&A, math & more.
Learn more
Gemini 3.5 Transcribe
Gemini 3.5 Transcribe is Google’s most precise speech-to-text model yet, designed for intelligent voice interactions and real-time transcription. Instead of simply converting speech word for word, it turns raw audio into accurate, polished, formatted text while handling background noise, complex jargon, accents, dialects, and natural speaking patterns. Smart transcription automatically understands self-corrections, removes filler words such as “ums” and “ahs,” and formats the final text for readability. The model supports continuous bidirectional streaming with sub-second latency for interactive voice applications, as well as pre-recorded audio processing for meetings, call logs, and other recordings with speaker attribution and word-level timestamps. Custom vocabulary helps it recognize specialized terminology, unique spellings, postal codes, order IDs, and other domain-specific language.
Learn more
Amazon Nova Sonic
Amazon Nova Sonic is a state-of-the-art speech-to-speech model that delivers real-time, human-like voice conversations with industry-leading price performance. It unifies speech understanding and generation into a single model, enabling developers to create natural, expressive conversational AI experiences with low latency. Nova Sonic adapts its responses based on the prosody of input speech, such as pace and timbre, resulting in more natural dialogue. It supports function calling and agentic workflows to interact with external services and APIs, including knowledge grounding with enterprise data using Retrieval-Augmented Generation (RAG). It provides robust speech understanding for American and British English across various speaking styles and acoustic conditions, with additional languages coming soon. Nova Sonic handles user interruptions gracefully without dropping conversational context and is robust to background noise.
Learn more
