Pinned Repositories
Fastspeech2_HS
Indic TTS for Indian Languages: This is a project on developing text-to-speech (TTS) synthesis systems for Indian languages, improving quality of synthesis, as well as small foot print TTS integrated with disability aids and various other applications.
Fastspeech2_HS_API
Fastspeech2_HS_with_Custom_dictionary
Inference the Fastspeech2_HS models using the custom dictionary. Some domain specific words gets mispronounced or some abbreviations are not pronounced as expected with the trained model. The custom dictionary can be used to alleviate the issue
Fastspeech2_MFA
Indic TTS for Indian Languages: This is a project on developing text-to-speech (TTS) synthesis systems for Indian languages, improving quality of synthesis, as well as small foot print TTS integrated with disability aids and various other applications.
generate_Lipsync_videos
This project generates a lip-synced video based on an original video and translated subtitles.
smtiitm.github.io
Training_Fastspeech2_HS_Model
This repository contains a recipe for training FastSpeech2 with Hybrid Segmentation (HS), a state-of-the-art text-to-speech (TTS) model. The training utilizes the Espnet toolkit and is tailored for the Indian languages, covering 13 major languages of India.
SMT lab 's Repositories
smtiitm/Fastspeech2_HS
Indic TTS for Indian Languages: This is a project on developing text-to-speech (TTS) synthesis systems for Indian languages, improving quality of synthesis, as well as small foot print TTS integrated with disability aids and various other applications.
smtiitm/Fastspeech2_MFA
Indic TTS for Indian Languages: This is a project on developing text-to-speech (TTS) synthesis systems for Indian languages, improving quality of synthesis, as well as small foot print TTS integrated with disability aids and various other applications.
smtiitm/generate_Lipsync_videos
This project generates a lip-synced video based on an original video and translated subtitles.
smtiitm/Fastspeech2_HS_with_Custom_dictionary
Inference the Fastspeech2_HS models using the custom dictionary. Some domain specific words gets mispronounced or some abbreviations are not pronounced as expected with the trained model. The custom dictionary can be used to alleviate the issue
smtiitm/Fastspeech2_HS_API
smtiitm/smtiitm.github.io
smtiitm/Training_Fastspeech2_HS_Model
This repository contains a recipe for training FastSpeech2 with Hybrid Segmentation (HS), a state-of-the-art text-to-speech (TTS) model. The training utilizes the Espnet toolkit and is tailored for the Indian languages, covering 13 major languages of India.