For speaker recognition, we need developement data to train T, UBM.
Is it possible that the single speech sample have more than 1 speaker i.e in
s = s1+s2+s3 ??
Dear Rizwan
I am not sure whether I understand you question properly. Is it about the availability of the database? If the answer is afirmative, Then if you mean convetsational dataset, it is available from LDC and you may check their catalogue at https://catalog.ldc.upenn.edu/ but it is not free.
If the question is about simulatenous speech, I am not sure. But you may generate that by combining two speech samples.
@Mohammadi, i.e. the single speech utterance can have multiple speakers, i.e. the single speech utterance is the combination of more 1 speakers for developement data???
I believe that in TIMIT database a single sentence pronounced by a number of speakers separately in different files, you may combine those samples for your application. Moreover, for conversational speech with two speakers LDC has several databases.
- Badges
- Science topic