Text-to-Speech
Natural speech synthesis in 18 languages, with female and male voices.
Speech Recognition
Multilingual speech recognition, transcription, and translation.
Audio-to-Audio
Fully interleaved voice chat with reasoning and generation in one model.
Audio Function Calling
Voice-driven tool use, agent workflows, and API orchestration.
LFM3 Models   Latest release
LFM3-Audio uses a new architecture with an audio encoder and codec developed in-house. It improves speech quality over LFM2.5-Audio, supports 18 languages, and adds audio understanding, general-purpose audio reasoning, and tool calling.
LFM3-Audio-2B
2B RecommendedMultilingual audio model for ASR, TTS, voice chat, and audio understanding.
LFM2.5 Models
LFM2.5-Audio adds a custom LFM-based audio detokenizer, llama.cpp-compatible GGUFs for CPU inference, and improved ASR and TTS performance over LFM2-Audio.
LFM2.5-Audio-1.5B-JP
1.5B JapaneseJapanese-focused audio model for ASR, TTS, and interleaved voice chat.
Guides
Learn how to use these models in your application:Audio Capabilities
Transcribe and translate speech, answer spoken questions, and ask about speech, music, and sounds.
Speech Generation
Text-to-speech, spoken chat, and spoken answers, with control over the output modality.
Tool Calling
Call functions from text, images, and speech.
Examples
Explore practical implementations using audio models:Real-Time Audio Transcription CLI
Transcribe audio files locally in real-time using LFM2.5-Audio-1.5B with llama.cpp for 100% private, on-device processing.Platform: Desktop · Uses: LFM2.5-Audio-1.5B
Browser Audio Demo with WebGPU
Run ASR, TTS, and interleaved conversations entirely in-browser using LFM2.5-Audio-1.5B with WebGPU acceleration.Platform: Web · Uses: LFM2.5-Audio-1.5B