VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
ସଠିକ ନିୟନ୍ତ୍ରଣ ପାଇଁ SSML ଟ୍ୟାଗଗୁଡ଼ିକରେ ଆପଣଙ୍କର ପାଠ୍ୟକୁ ଲଗାନ୍ତୁ:
<speak><prosody rate="slow">Slow speech</prosody></speak>
ବଚ୍ଛିତ ନମୁନା ବୁଝିଥିବା ସୂଚକଗୁଡ଼ିକ - ଏହା ଘଟୁଥିବା ସ୍ଥାନକୁ ଆପଣଙ୍କର ପାଠ୍ୟରେ ଗୋଟିଏ ପକାଇବା ପାଇଁ କ୍ଲିକ କରନ୍ତୁ:
ଏହି ଆକାର ସରଳ ପାଠ୍ୟ ପଢ଼େ, ତେଣୁ ଅନ୍ତର୍ନିହିତ ସୂଚକଗୁଡ଼ିକୁ ଅଣଦେଖା କରାଯାଏ। ସୂଚକ ଆଧାରିତ ଅନୁଭୂତି ପାଇଁ, ଗୋଟିଏ ଅଭିବ୍ୟକ୍ତିମୂଳକ ଆକାରକୁ ପରିବର୍ତ୍ତନ କରନ୍ତୁ ଯେପରିକି Orpheus କିମ୍ବା Bark।
ଇଚ୍ଛାରୂପୀ ଉଚ୍ଚାରଣକୁ ବର୍ଣ୍ଣନା କରନ୍ତୁ (ଶବ୍ଦ = ଉଚ୍ଚାରଣ):
ବିଷୟରେ VITS
VITS — Variational Inference with adversarial learning for end-to-end Text-to-Speech — was introduced by Jaehyeon Kim and collaborators in 2021 and became a foundational architecture for modern neural speech. Rather than the older two-stage pipeline, it synthesizes audio in a single parallel end-to-end pass, pairing a variational autoencoder with normalizing flows and a GAN-style adversarial training process to lift naturalness. At about 25M parameters and trained on ~585 hours, it produces natural prosody at fast inference speeds and supports multiple speakers. It serves as a solid general-purpose, free baseline and underpins many later models such as Piper and MeloTTS.
ପାଇଁ ଉତ୍ତମ: General-purpose text-to-speech with natural prosody
ସମସ୍ତଙ୍କୁ ବ୍ରାଉଜ କରନ୍ତୁ VITS ଧ୍ୱନିଗୋଟିଏ ନଜରରେ
- ବିକାଶକାରୀ
- Jaehyeon Kim et al.
- ଅନୁମତିପତ୍ର
- MIT
- ତିଆର
- free
- ବେଗ
- fast
- ଧ୍ୱନି କ୍ଲୋନିଂ
- ନାମ
- ଭାଷାName
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- ସର୍ବାଧିକ ଅକ୍ଷର
- 2000