Automatic Speech Recognition
Transformers
Safetensors
Chinese
English
audio8_asr_infinite
text-generation
streaming
realtime
speech-recognition
audio
custom_code
Instructions to use Edge0/Audio8-ASR-Infinite with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Edge0/Audio8-ASR-Infinite with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Edge0/Audio8-ASR-Infinite", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("Edge0/Audio8-ASR-Infinite", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "architectures": [ | |
| "Audio8ASRInfiniteForConditionalGeneration" | |
| ], | |
| "audio_config": { | |
| "activation_function": "gelu", | |
| "attention_dropout": 0.0, | |
| "dtype": "bfloat16", | |
| "head_dim": 64, | |
| "hidden_act": "silu", | |
| "hidden_size": 1280, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 5120, | |
| "max_position_embeddings": 1500, | |
| "model_type": "voxtral_realtime_encoder", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_mel_bins": 128, | |
| "rms_norm_eps": 1e-05, | |
| "rope_parameters": { | |
| "rope_theta": 1000000.0, | |
| "rope_type": "default" | |
| }, | |
| "sliding_window": 750, | |
| "streaming_n_left_pad_tokens": 18, | |
| "vocab_size": 131072 | |
| }, | |
| "audio_length_per_tok": 8, | |
| "audio_tower_frame_ms": 20, | |
| "auto_map": { | |
| "AutoConfig": "configuration_audio8_asr_infinite.Audio8ASRInfiniteConfig", | |
| "AutoModelForCausalLM": "modeling_audio8_asr_infinite.Audio8ASRInfiniteForConditionalGeneration" | |
| }, | |
| "bos_token_id": 151644, | |
| "default_num_delay_tokens": null, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 151645, | |
| "frame_lens": [ | |
| 4, | |
| 6, | |
| 8 | |
| ], | |
| "hidden_size": 2048, | |
| "max_frame_len": 8, | |
| "model_type": "audio8_asr_infinite", | |
| "num_delay_tokens_by_frame_len": { | |
| "4": { | |
| "240": 3, | |
| "320": 4, | |
| "480": 6, | |
| "560": 7 | |
| }, | |
| "6": { | |
| "240": 2, | |
| "480": 4 | |
| }, | |
| "8": { | |
| "320": 2, | |
| "480": 3 | |
| } | |
| }, | |
| "pad_token_id": 151643, | |
| "processor_class": "VoxtralRealtimeFeatureExtractor", | |
| "projection_size": 10240, | |
| "projector_hidden_act": "gelu", | |
| "semantic_vad_horizons_seconds": [ | |
| 0.5, | |
| 1.0, | |
| 2.0, | |
| 3.0 | |
| ], | |
| "semantic_vad_num_classes": 8, | |
| "streaming_frame_ms": null, | |
| "streaming_n_left_pad_tokens": 18, | |
| "streaming_n_left_pad_tokens_by_frame_len": { | |
| "4": 18, | |
| "6": 12, | |
| "8": 9 | |
| }, | |
| "supported_frame_lens": [ | |
| 4, | |
| 6, | |
| 8 | |
| ], | |
| "target_delay_ms": [ | |
| 240, | |
| 320, | |
| 480, | |
| 560 | |
| ], | |
| "text_config": { | |
| "_name_or_path": "Qwen/Qwen2.5-3B-Instruct", | |
| "architectures": [ | |
| "Qwen2ForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151644, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 151645, | |
| "hidden_act": "silu", | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 11008, | |
| "layer_types": [ | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention" | |
| ], | |
| "max_position_embeddings": 32768, | |
| "max_window_layers": 70, | |
| "model_type": "qwen2", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 36, | |
| "num_key_value_heads": 2, | |
| "pad_token_id": 151643, | |
| "projection_size": 10240, | |
| "rms_norm_eps": 1e-06, | |
| "rope_parameters": { | |
| "rope_theta": 1000000.0, | |
| "rope_type": "default" | |
| }, | |
| "sliding_window": null, | |
| "tie_word_embeddings": true, | |
| "use_cache": false, | |
| "use_sliding_window": false, | |
| "vocab_size": 151936 | |
| }, | |
| "tie_word_embeddings": true, | |
| "token_duration_ms": null, | |
| "tokenizer_class": "Qwen2Tokenizer", | |
| "transformers_version": "5.13.0", | |
| "use_cache": false, | |
| "use_frame_len_embedding": true, | |
| "vocab_size": 151936, | |
| "weight_format_version": 2 | |
| } | |