172 Commits

Author	SHA1	Message	Date
嘉渊	6e66a74ae6	update	2023-04-25 16:33:00 +08:00
嘉渊	7436acc5dd	update	2023-04-25 16:29:39 +08:00
嘉渊	70f9a8f890	update	2023-04-25 01:29:12 +08:00
嘉渊	e86b95e747	update	2023-04-24 22:57:04 +08:00
嘉渊	f2b9780b29	update	2023-04-24 22:47:00 +08:00
haoneng.lhn	a8e92e4fb4	update data filtering recipe	2023-04-23 15:03:56 +08:00
speech_asr	993fdd8ecf	update	2023-04-20 17:01:47 +08:00
speech_asr	eac9f111b5	update	2023-04-20 16:59:26 +08:00
speech_asr	3e77fd4430	update	2023-04-20 16:41:22 +08:00
speech_asr	d6cc6896e4	update	2023-04-20 16:33:30 +08:00
speech_asr	518465d089	update	2023-04-20 16:07:01 +08:00
speech_asr	a29166b9a0	update	2023-04-20 16:03:54 +08:00
speech_asr	200d1ede05	update	2023-04-20 15:56:25 +08:00
speech_asr	c452b2a3c7	update	2023-04-20 15:43:29 +08:00
speech_asr	68852c3072	update	2023-04-20 15:35:25 +08:00
speech_asr	43c30967b0	update	2023-04-20 11:48:19 +08:00
speech_asr	02f2a3c2ec	update	2023-04-20 11:38:20 +08:00
speech_asr	7522c59e74	update	2023-04-20 11:16:49 +08:00
speech_asr	680cdb55bb	update	2023-04-19 14:49:36 +08:00
speech_asr	58fb22cb2b	update	2023-04-19 10:09:51 +08:00
speech_asr	05d4176e88	update	2023-04-18 19:28:33 +08:00
speech_asr	831d00aec2	update	2023-04-17 16:26:40 +08:00
speech_asr	d9ad40bf6f	update	2023-04-17 11:45:41 +08:00
speech_asr	6659c37d81	update	2023-04-17 11:23:37 +08:00
speech_asr	bd7455ec7d	update	2023-04-12 10:43:01 +08:00
北念	cf843d144a	fix compute cer problems	2023-04-04 14:26:22 +08:00
hnluo	85e8e0ed0d	Update postprocess_utils.py	2023-03-27 17:15:10 +08:00
Xian Shi	92248eb07b	Merge pull request #274 from dingbig/fix-punc ... fixed token_int is zero bug and add more puncs to sentence	2023-03-21 17:52:18 +08:00
Yuanhang Zhang	3b3aebb124	Fix timestamp prediction on empty ASR outputs ... _, timestamp = ts_prediction_lfr6_standard(us_alphas[i], ValueError: not enough values to unpack (expected 2, got 0) If char_list is empty, we should still return both the text (which is empty string) and the timestamps to fit the function signature.	2023-03-21 17:23:26 +08:00
dingbig	d80d046365	fixed token_int is zero bug and add more puncs to sentence ... 1)If token_int is 0, the following process will crash. This BUG has been fixed. In fact, when token_int=0, there is no need to continue processing. 2)modify time_stamp_sentence to support more punc.	2023-03-21 17:11:49 +08:00
lzr265946	2ab8b7f473	modify abbr postprocess	2023-03-21 12:55:16 +08:00
仁迷	60c8f036e0	update audio type check	2023-03-17 20:09:02 +08:00
shixian.shi	f63a72c52e	update tools	2023-03-15 10:22:30 +08:00
shixian.shi	f59a72d24e	release timestasmp related tools	2023-03-15 10:21:32 +08:00
shixian.shi	0b06794fde	bug fix	2023-03-13 20:14:41 +08:00
shixian.shi	9c21bbb96b	bug fix	2023-03-13 19:55:47 +08:00
shixian.shi	4b16316d49	bug fic	2023-03-13 19:53:33 +08:00
shixian.shi	71d466e745	update AverageShiftCalculator in utils	2023-03-13 19:47:42 +08:00
shixian.shi	f5aa97f7bf	update params name	2023-03-13 17:39:18 +08:00
shixian.shi	5a7ee30783	update timestamp related codes and egs_modelscope	2023-03-13 15:21:13 +08:00
zhifu gao	0a09368a64	Merge pull request #180 from zhuzizyf/finetune_fix ... Update wav_utils.py	2023-03-03 11:01:21 +08:00
zhuzizyf	1a39b6f981	Update wav_utils.py ... Because there are no uppercase letters in the dictionary, when there are uppercase letters in the annotated text, the finetune result will be "unk", so uniformly converted to lowercase when read the annotated text.	2023-03-03 10:33:51 +08:00
shixian.shi	9dd4901aad	rapid_paraformer.utils.timestamp_utils	2023-03-02 19:50:06 +08:00
zhifu gao	5d4b0c3994	Merge pull request #167 from alibaba-damo-academy/dev_lhn ... fix text postprocess bug	2023-03-01 11:17:18 +08:00
仁迷	e9ea65679a	fix text postprocess bug	2023-03-01 11:14:09 +08:00
shixian.shi	57f2a51f9a	onnx supports tiny and bicif paraformer	2023-02-27 16:55:06 +08:00
仁迷	b6a1c6c1e6	fix data dir filter bug	2023-02-27 14:47:32 +08:00
dingbig	bea5d98423	Add sentence timestamp support ... Added support for statement event timestamp, which is particularly useful for applications such as lyrics and subtitles.	2023-02-22 19:48:50 +08:00
shixian.shi	03250ae634	timestamp func bug fix	2023-02-21 10:10:17 +08:00
hnluo	4a796298cc	Update asr_utils.py	2023-02-20 15:01:50 +08:00
lzr265946	546262a0c6	remove useless code	2023-02-16 15:22:14 +08:00
hnluo	71766839fd	Merge pull request #106 from alibaba-damo-academy/dev ... Dev	2023-02-14 15:15:15 +08:00
speech_asr	216dc0978c	add wav/text mismatch process	2023-02-14 14:58:46 +08:00
speech_asr	66a8235fbf	add wav/text mismatch process	2023-02-14 14:45:39 +08:00
speech_asr	e180abe4fd	update docs	2023-02-14 14:39:45 +08:00
志浩	f6a1cdaf34	add sond model	2023-02-10 18:56:14 +08:00
lzr265946	7aa2e885f4	support for turning off timestamps	2023-02-10 13:46:01 +08:00
北念	ad0039596c	add BiCifParaformer	2023-02-09 19:11:16 +08:00
北念	16d4e00549	add BiCifParaformer	2023-02-09 17:53:04 +08:00
hnluo	d7e43300fb	Merge pull request #81 from alibaba-damo-academy/dev ... Create vad_inference_launch.py	2023-02-09 15:30:21 +08:00
lzr265946	03875965c8	remove global vars	2023-02-09 15:13:14 +08:00
speech_asr	cced441e5f	add file flush	2023-02-09 15:12:14 +08:00
zhifu gao	08384ef9eb	Merge pull request #75 from alibaba-damo-academy/dev ... update github.io page	2023-02-08 19:52:18 +08:00
jmwang66	6a41e13ba9	upload github.io	2023-02-08 17:30:19 +08:00
hnluo	02a3eefb35	fix audio_in type bug	2023-02-07 10:09:51 +08:00
hnluo	9c64377c98	support pcm audio format	2023-02-06 17:05:06 +08:00
仁迷	cdac560080	more audio formats support	2023-01-31 17:42:28 +08:00
jmwang66	12a7adfdf3	update version 0.1.6	2023-01-16 18:46:40 +08:00
仁迷	ad75a464d8	update text postprocess	2023-01-09 19:34:52 +08:00
jmwang66	d653df71cb	update text postprocess	2022-12-23 10:40:23 +08:00
lzr265946	a9e857e452	update funasr 0.1.3	2022-12-03 16:39:38 +08:00
游雁	c087854f71	create	2022-11-26 21:56:51 +08:00