Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

分享自用参数(对于没有杂音和音乐的网课类英语音频近乎完美) #247

Open
czyrichard opened this issue Nov 7, 2024 · 0 comments

Comments

@czyrichard
Copy link

1)使用的模型:faster-whisper-large-v3-turbo-ct2
2)建议每次针对音频的内容对hotwords进行更改,对提升转写质量有奇效。

{
"theme": "light",
"demucs": {
"overlap": 0.1,
"segment": 10.0,
"tracks": 1
},
"model_param": {
"localModel": true,
"onlineModel": false,
"modelName": 0,
"use_v3_model": true,
"device": 1,
"deviceIndex": "0",
"preciese": 4,
"thread_num": "4",
"num_worker": "1",
"local_files_only": true
},
"vad_param": {
"use_VAD": false,
"threshold": 0.4,
"minSpeechDuration": "250",
"minSilenceDuration": "100",
"maxSpeechDuration": "30",
"windowSize": 2,
"speechPad": "2000"
},
"setting": {
"saveConfig": true,
"autoLoadModel": false,
"language": 1,
"autoGoToOutputPage": 2,
"autoClearTempFiles": false,
},
"Transcription_param": {
"aggregate_contents": false,
"language": 1,
"task": false,
"beam_size": "10",
"best_of": "5",
"patience": "2",
"length_penalty": "3",
"temperature": "0.0,0.2,0.4,0.6,0.8,1.0",
"compression_ratio_threshold": "5",
"log_prob_threshold": "-1.0",
"no_speech_threshold": "0.2",
"condition_on_previous_text": true,
"initial_prompt": "Hello.",
"prefix": "",
"suppress_blank": true,
"suppress_tokens": "-1",
"without_timestamps": false,
"max_initial_timestamp": "9999999.0",
"word_timestamps": false,
"prepend_punctuations": ""'“¿([{-",
"append_punctuations": ""'.。,,!!??::”)]}、",
"repetition_penalty": "1",
"no_repeat_ngram_size": "0",
"prompt_reset_on_temperature": "0.5",
"chunk_length": "30",
"clip_mode": 0,
"max_new_tokens": "448",
"clip_timestamps": "",
"hallucination_silence_threshold": "0",
"hotwords": "This is a web seminar about the placement of American English accent.",
"language_detection_threshold": "",
"language_detection_segments": "1"
},
"output_whisperX": {
"tabMovable": true,
"tabScrollable": false,
"tabShadowEnabled": false,
"tabMaxWidth": 500,
"closeDisplayMode": 0,
"whisperXMinSpeaker": 0,
"whisperXMaxSpeaker": 0,
"outputFormat": 5,
"outputEncoding": 1
}
}

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant