FireRedASR
FireRedASR copied to clipboard
FireRedTeam
Open-source industrial-grade ASR models supporting Mandarin, Chinese dialects and English, achieving a new SOTA on public Mandarin ASR benchmarks, while also offering outstanding singing lyrics recogn...
For long audio (e.g.,> 20 min), what is the best practice? What do you think of the following method? - Split into 30s chunks, with 10s overlap. - Get a...
在使用torch==2.6.0的时候，运行inference_fireredasr_aed.sh 会报错，需要稍微降低一下版本 ``` Traceback (most recent call last): File "/home/orange/FireRedASR/examples/fireredasr/speech2text.py", line 105, in main(args) File "/home/orange/FireRedASR/examples/fireredasr/speech2text.py", line 43, in main model = FireRedAsr.from_pretrained(args.asr_type, args.model_dir) File "/home/orange/FireRedASR/examples/fireredasr/models/fireredasr.py", line 25, in from_pretrained...
huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name'
fixes error: ```shell torch/serialization.py", line 1470, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options,...
在 `asr_feat.py` 中， 1. `frame_length`, `frame_shift` 被传入了 `KaldifeatFbank` 但并未被使用 2. `wav` 中的 `sample_rate` 也没有传入 knf，导致所有音频都按照 16k 采样率提取特征 ```python class KaldifeatFbank: def __init__(self, num_mel_bins=80, frame_length=25, frame_shift=10, dither=1.0): self.dither = dither opts...
may ask if a sketch of hyperparameters can be shown.
def speech_to_text(audio_path): logger.debug("开始语音识别") # 初始化 FireRedASR 模型 model = FireRedAsr.from_pretrained("aed", "pretrained_models/FireRedASR-AED-L") 执行报错 File "D:\AI\FireRedASR\app.py", line 83, in main result = speech_to_text(audio_path) File "D:\AI\FireRedASR\app.py", line 17, in speech_to_text model = FireRedAsr.from_pretrained("aed",...
It is a great work. But in my experience, it is not as good as gpt-4o-transcribe even for Chinese.
Metadata
Owner
Metadata
Open-source industrial-grade ASR models supporting Mandarin, Chinese dialects and English, achieving a new SOTA on public Mandarin ASR benchmarks, while also offering outstanding singing lyrics recogn...