VibeVoice-ASR (7B), the original BF16 weights as Microsoft published them, with its DFlash 2 drafter bundled in drafter/: one download, and vibevoice.c decodes with speculative decoding -- the drafter proposes 8 tokens in one pass, the model checks them in one pass and keeps the ones it agrees with. the model's own decode step, so the transcript is byte-for-byte the one without the drafter. (With a drafter vibevoice.c attends with flashinfer, which checks rows together; the transcript is then --attn flashinfer's.) Needs vibevoice.c with DFlash 2 support: branch dflash2 (PR #48), in the next release. A model directory's drafter/ is used without asking: vibevoice.c runs BF16 weights as dense…
Open weights
mit
8.7B parameters
vibevoice.c
VibeVoice-ASR (7B), AWQ W4A16 (asymmetric, groups of 128), with its DFlash 2 drafter bundled in drafter/: one download, and vibevoice.c decodes with speculative decoding -- the drafter proposes 8 tokens in one pass, the model checks them in one pass and keeps the ones it agrees with. the model's own decode step, so the transcript is byte-for-byte the one without the drafter. (With a drafter vibevoice.c attends with flashinfer, which checks rows together; the transcript is then --attn flashinfer's.) Needs vibevoice.c with DFlash 2 support: branch dflash2 (PR #48), in the next release. A model directory's drafter/ is used without asking: vibevoice.c b72be15 (branch dflash2), RTX 3090, greedy…
Open weights
mit
8.3B parameters
vibevoice.c
Voxtral Mini 4B Realtime 2602 is a multilingual, realtime speech-transcription model and among the first open-source solutions to achieve accuracy comparable to offline systems with a delay of = 3600 / 0.8 = 45000. In theory, you should be able to record with no limit; in practice, pre-allocations of RoPE parameters among other things limits --max-model-len. For the best user experience, we recommend to simply instantiate vLLM with the default parameters which will automatically set a maximum model length of 131072 (~ca. 3h).
Open weights
apache-2.0
4.4B parameters
131,072 tokens
vllm
Model · Speech recognition
Edge0
Audio8 ASR Infinite is a native streaming speech recognition model built to be as responsive as possible. It offers a selectable audio clock (80/120/160 ms) and a transcription delay (240–560 ms). With our adapted vLLM build it transcribes unlimited-length audio 24/7 without drifting. - Super responsive — the native streaming architecture decodes 12.5 times per second. - Unlimited-length transcription — a rolling KV Cache keeps both memory and latency constant, even in 24/7 operation. - Selectable streaming clock — one text token per clock step (12.5 / 8.3 / 6.25 decisions per second), balancing perception granularity and resource cost. - Configurable transcription delay — set how much…
Open weights
apache-2.0
4.1B parameters
32,768 tokens
transformers
MLX-native int8 conversion of Qwen3-ASR-1.7B for local automatic speech recognition on Apple Silicon — English, Chinese, and mixed Chinese/English. It runs through the mlx-speech runtime with no PyTorch and no cloud API at inference time. Weights ship as plain.safetensors. Install mlx-speech, then load by alias (downloads on first use): Or download the weights once and load by local path: - This repo contains the quantized MLX runtime artifact only. - The quantization mode (affine, groupsize 64, 8-bit) is stored in config.json and re-applied automatically on load, so it cannot desync from the weights. - v0 is single-pass offline ASR; streaming, timestamps, and long-audio chunking are…
Open weights
apache-2.0
2.4B parameters
mlx
Model · Speech recognition
Qwen
The Qwen3-ASR family includes Qwen3-ASR-1.7B and Qwen3-ASR-0.6B, which support language identification and ASR for 52 languages and dialects. Both leverage large-scale speech training data and the strong audio understanding capability of their foundation model, Qwen3-Omni. Experiments show that the 1.7B version achieves state-of-the-art performance among open-source ASR models and is competitive with the strongest proprietary commercial APIs. Here are the main features: Novel and strong forced alignment Solution: We introduce Qwen3-ForcedAligner-0.6B, which supports timestamp prediction for arbitrary units within up to 5 minutes of speech in 11 languages. Evaluations show its timestamp…
Open weights
apache-2.0
2.3B parameters
