From 6e69d784e4814c3dbe35e8f70c6cf4b920c8b20b Mon Sep 17 00:00:00 2001
From: 天地 <tiandiweizun@gmail.com>
Date: 星期三, 19 三月 2025 23:10:13 +0800
Subject: [PATCH] 1. bug fix:list(mean)和list(var)，由于mean和var是numpy，导致写入到文件的格式错误，参考上面的话，大概率是list(mean.tolist())，其实外层list没有必要 (#2437)
---
funasr/models/whisper/template.yaml | 31 +++++++++++++++----------------
1 files changed, 15 insertions(+), 16 deletions(-)
diff --git a/funasr/models/whisper/template.yaml b/funasr/models/whisper/template.yaml
index 40b902c..568b948 100644
--- a/funasr/models/whisper/template.yaml
+++ b/funasr/models/whisper/template.yaml
@@ -11,29 +11,28 @@
lsm_weight: 0.1
length_normalized_loss: true
hub: funasr # openai
- init_param_path: null # large-v2 or large-v3 if hub == "openai"
# only use for hub == funasr,
-# if hub == openai, whisper_dims is automaticall download
-whisper_dims:
- 'n_mels': 80
- 'n_vocab': 51865
- 'n_audio_ctx': 1500
- 'n_audio_state': 1280
- 'n_audio_head': 20
- 'n_audio_layer': 32
- 'n_text_ctx': 448
- 'n_text_state': 1280
- 'n_text_head': 20
- 'n_text_layer': 32
+# if hub == openai, dims is automaticall download
+dims:
+ n_mels: 128
+ n_vocab: 51866
+ n_audio_ctx: 1500
+ n_audio_state: 1280
+ n_audio_head: 20
+ n_audio_layer: 32
+ n_text_ctx: 448
+ n_text_state: 1280
+ n_text_head: 20
+ n_text_layer: 32
# frontend related
frontend: WhisperFrontend
frontend_conf:
fs: 16000
- n_mels: 80
+ n_mels: ${dims.n_mels}
do_pad_trim: true
tokenizer: WhisperTokenizer
@@ -41,6 +40,6 @@
language: null
task: transcribe
is_multilingual: true
- num_languages: 99
+ num_languages: 100
-scope_map: ['none', "model."]
+scope_map: [none, "model."]
\ No newline at end of file
--
Gitblit v1.9.1