Open Cloud Speech to Text Issues

1 - 50 of 224

Saved search: Open Cloud Speech to Text Issues

ID	TYPE	STATUS	VOTES	TITLE	P	S	ASSIGNEE	BLOCKED BY
307574382	Bug	Assigned	20	Google Speech To Text repeats short mono syllabic words more often than present in audio	P2	S2		310817064
346262110	Feature Request	New	11	PL language in chirp_telephony	P2	S2	--	346268563
336958062	Feature Request	Assigned	9	Japanese regional dialects not being recognised well by Speech to Text	P2	S2		336955567
197266339	Feature Request	Assigned	8	Implementing Speak Diarization for Swedish language	P2	S2		197185078
197265839	Feature Request	Assigned	7	Implementing Phone Call model for Swedish language	P2	S2		197184876
328499488	Customer Issue	Assigned	6	[Speech-to-Text V2] CXR Phrase is not working well with long model	P2	S2		327254027
199112218	Bug	Assigned	5	Speaker tag numbers are out of order and Speaker diarization is very inaccurate.	P2	S2		200008851
170653935	Bug	Assigned	5	Dialogflow timeout using small input phrases as "sí" or "no" using Spanish language	P2	S2		135267079, 148222884
318671811	Bug	Assigned	4	Invalid Argument when certain characters are included in InlinePhraseSet	P2	S2
234121595	Feature Request	Assigned	4	Speech to text not able to recognize a series of 1s	P2	S2		260044685
218160601	Bug	Assigned	4	Outdated Hebrew language code	P2	S2		218500219
192771764	Feature Request	Assigned	4	Add returning auto-selected model for SpeechRecognitionResult	P2	S2		193209439
370627603	Feature Request	Assigned	3	SPEECH_ACTIVITY_END event not returned for Japanese audio in long model.	P2	S2		370762809
335055223	Feature Request	Assigned	3	Cx can not get "Katakana"(カタカナ) notation for person's name after upgrading to the Conformer model on V1 API, Speech To Text.	P2	S2		330841091
330443171	Bug	Assigned	3	In multiple languages, result.IsFinal is no longer processed.	P2	S2
308746327	Bug	Assigned	3	Quality issues with Covid terms for medical_conversation STT model	P2	S2		302297892
242217847	Customer Issue	Assigned	3	Empty transcripts for nl-nl language with dual channel recording	P2	S2		236880359
214094971	Customer Issue	New	3	STT API not transcribing short duration phone audio converted from G.711 Alaw to LINEAR16 using lookup table, for English (UK) language	P2	S2	--	209064572
396175593	Feature Request	Assigned	2	Different Endpoints using the same transcription for the telephony model bring different results	P2	S2		392863554
378421882	Bug	Assigned	2	Interim results were requested, but the chirp_2 model does not support it (hebrew)	P2	S2
364414921	Feature Request	Assigned	2	Insert spaces in Japanese phrases (文節)	P2	S2		364416018
364026452	Feature	New	2	Dialogflow CX - Special Characters Not Recognized in Voice Input	P2	S2	--	364028666
362883814	Feature Request	Assigned	2	Speech-to-text v2 - Extend the support for streaming recognize for "chirp_2" model.	P2	S2		362689363
362409028	Feature Request	Assigned	2	Lack of streaming speech-to-text examples for Speech-To-Text API V2	P2	S2		332484976
361493635	Bug	New	2	Speech-to-Text API: Internal Errors with Chirp Models for Chinese Transcription	P2	S2	--	357782915
354262888	Feature Request	Assigned	2	Transcription of chirp 2 presents a bug with repetitions of words such as "é é é é é é....	P2	S2		351814052
353631787	Feature Request	New	2	Request to exclude Japanese filler words from Speech-to-Text	P2	S2	--	353631321
352033315	Feature Request	Assigned	2	chirp_2 model to support word_timestamps/Time offsets	P2	S2		352033479, 345253088
351148341	Feature Request	Assigned	2	Recognition issues with Polish lang, short model gsr@v2	P2	S2		351146218
349753183	Customer Issue	Assigned	2	Issue recognizing repeated digits sequences in Polish Language	P2	S2		346256235
338135243	Bug	New	2	Can't migrate to conformer models due to performance degradation	P2	S2	--	327258348
337766123	Bug	Assigned	2	Get latency for realtime speech to text of around 2 seconds.	P2	S2
337310124	Feature Request	Assigned	2	Multi-language support for Speech-to-Text (Streaming) in a single utterance/audio.	P3	S2		337310104
335777549	Bug	Assigned	2	Speech to text voice recognition for Japanese phone number is poor and quality of voice recognition should be improvised.	P2	S2
331358159	Feature Request	Assigned	2	Request for Somali Language Support in Google Cloud Speech-to-Text*	P2	S2
328375564	Feature Request	New	2	Adding ABNF grammar at Cloud Speech-to-Text V2	P2	S2	--	280044324
307362959	Feature Request	Assigned	2	Option to have Speech-to-Text API Request Results to Appear in Cloud Console UI	P2	S2		307366330
306517539	Feature Request	Assigned	2	The customer has some question regarding the stability score when using streaming response of Speech-to Text API	P2	S2
275157236	Bug	Assigned	2	with model=latest_short and single_uttterance=false I can get an end of utterance event and then nothing else	P2	S2		277128887
229654786	Feature Request	Assigned	2	Automatic punctuation support for Ukrainian (Ukraine) uk-UA	P4	S4		228497121
211920225	Feature Request	New	2	Adding Traditional Chinese(zh-tw) support for enhanced phone call mode	P1	S2	--	190476344
202760222	Customer Issue	Assigned	2	Indian English model does not perform well in far-field scenario	P2	S2		207768287
182892089	Feature Request	Assigned	2	Customer would like to have Polish language added in Speaker diarization.	P2	S2		182892378
156636175	Feature Request	New	2	Modify C++ StreamingRecognize sample code to support channel pooling	P2	S2	--	156477522
143468770	Bug	New	2	Top alternative does not contain a words list for INTERIM results with speaker diarization enabled	P2	S2	--	149370507
137876113	Feature Request	Assigned	2	Create new metric to monitor Speech to text API audio seconds usage in a day	P2	S2		137759687
391255302	Feature Request	Assigned	1	[Feature request] Option to distinguish certain types of numbers in ja-JP	P2	S2		391252942
388272896	Feature Request	Assigned	1	Request for Chirp_2 model to fully support Devanagari notation for hi-IN.	P2	S2		373766001, 389791697
373478547	Bug	Assigned	1	Chirp 2 hallucinating conspiracy theories while transcribing Slovak audio	P2	S2
368241680	Bug	Assigned	1	Speech-to-Text Fails to Capture Most Lyrics in Song Transcriptions	P2	S2