Tecnologias de fala para russo
Comunicação
- https://t.me/speech_recognition_ru - Grupo "Reconhecimento de fala"
- https://t.me/speech_recognition - Grupo em inglês
- https://t.me/speechtech - canal de notícias
- https://t.me/betterdatacommunity/15 - Comunidade de fala em Datacommunity
- https://t.me/voitStestuff https://t.me/voice_stuff_chat - Espaço de Frappucino
- https://t.me/teraspace https://t.me/teraspace_chat - Espaço de Tera
Cursos
- https://github.com/markovka17/dla
- https://github.com/yandexdataSchool/speech_course
- https://github.com/severilov/dl- áudio-course
- https://hugingface.co/learn/audio-couurse/en/chapter0/introduction - brinque com modelos de som hf
- https://www.youtube.com/playlist?list=plyg3whdp5cwvrxljxzblqiqtwy_qjkmz - Deep Learning for Audio
Dados para treinamento
- https://github.com/salute-developers/golos
- https://github.com/snakers4/open_stt
- https://github.com/georgefedoseev/deepSpeech
- https://github.com/sovaai/sova-dataset
- https://www.openslr.org/96/ - russo Librispeech
- https://commonvoice.mozilla.org/ru/datasets - MCV
Síntese de fala
- https://www.caito.de/2019/01/03/the--m-ailabs-sech-dataset/-m-ailabs DataSet (da Librivox)
- https://ruslan-corpus.github.io/
- https://github.com/sovaai/sova-tts
- https://hugingface.co/bene-ges/tts_ru_hifigan_ruslan
- https://github.com/alphacep/vosk-tts
- https://github.com/rhvoice
- https://github.com/snakers4/silero-models#text-tonce
- https://github.com/tera2space/teratts
- https://hugingface.co/mog/xts-ru-ipa
Transformações de voz
- https://www.weights.gg/ru - um monte de modelos para RVC
- https: // 2ch- ai.gitgud.site/wiki/speech/ - Escola e Dual
- https://lunaiproject.uwu.ai/ - Russian Diffsinger
- Existem um monte de canais de telegrama, principalmente orientação enlameada
Contagem geral de síntese
- https://github.com/sovaai/sova-tts-tps
- https://github.com/snakers4/silero-models#text-enhancement
- https://github.com/snakers4/russian_stt_text_normalization
- https://www.kagghe.com/competitions/text-normalization-prussian-laguage/verview Old Competition for kaggle
- https://github.com/ppleskov/text-normalization-challenge-russian-laguage-one dos vencedores
- https://github.com/shigabeev/russian_tts_normalization
- https://github.com/saarus72/text_normalization/tree/dev - Baseado no FRED -T5
- https://github.com/den4ika/runorm - números em texto, processando palavras em inglês, divulgação de abreviações
- https://github.com/Just-Am-A/Multilinguly-Text-parser
Estresse de aperto, dicionários fonéticos e G2P
- https://github.com/reynoldsnlp/udar
- https://github.com/einhornus/russian_acentuation
- https://github.com/wilpert/rusphonetizer
- https://hugingface.co/bene-ges/ru_g2p_ipa_bert_large
- https://github.com/desklop/stressrnn
- https://github.com/nsu- arussian_g2p
- https://github.com/nsu- ai- equipe/russian_g2p_neuro
- https://github.com/suralmasha/rutranscript
- https://github.com/mashapo/russtress
- https://hugingface.co/ilyagusev/ru-word-tress-transformer
- https://github.com/aishutin/rustress
- https://github.com/koziev/stressmodel
- https://github.com/mogr/omogre
- https://github.com/den4ika/ruaccent - Yelifier, estresse e resolução de homógrafos
Dicionário
- https://github.com/reynoldsnlp/udar/blob/src/src/resources/src/tixonov.txt - Morphematum -orfographic Dictionary of Tikhonov
- http://aot.ru - a fonte do dicionário do pântano no formato da máquina
- https://github.com/gramdict/gramdict - Uma versão moderna
- http://odict.ru/ - outro desenvolvimento
- http://opencorpora.org/ - Dicionário morfológico marcado
- https://ru.wiktiony.org - Wikcionário
- https://kaikki.org/dictionary/russian/ - dump wikcionaly em um formato conveniente
Yeephores
- https://github.com/sovaai/sova-tts-tps
- https://github.com/e2yo/eyo-cernel
- https://github.com/kalashnikovisme/karamzin
- https://github.com/text-extend-tools/python-yficator
- https://github.com/emacsmiror/yoficator
- https://github.com/unabashed/yoficator
Reconhecimento de emoções
- https://github.com/aniemore/aniemore
- https://hugingface.co/xbgoose/hubert-lage-speech-motion-recognition-russian-dusha-finetified
- https://github.com/salute-developers/golos/tree/master/dusha
Modelos de reconhecimento de fala
Comparação de modelos aqui.
- Vosk pequeno https://alphacephei.com/vosk/models/vosk-model-small-ru-0.22.zip
- Vosk Big 0.22 https://alphacephei.com/vosk/models/vosk-model-0.22.zip
- Vosk Big 0.42 https://alphacephei.com/vosk/models/vosk-model-0.42.zip
- Nvidia rnnt https://hugingface.co/nvidia/stt_ru_conformer_transducer_large
- Whisper Medium https://github.com/openai/whisper
- Whisper Adapted Medium https://hugingface.co/mitcheldehaven/whisper-medium-ru
- Whisper adaptou Https://hugingface.co/mbitcheldehaven/whisper-large-v2-u2-
- Wav2veclm https://hugingface.co/jonatasgrosman/wav2vec2-xls-r-1b-russian
- Wav2veclm bond005 https://hugingface.co/bond005/wav2vec2-large-ru-golos (versão 03.2023)
- Saudar Citrinet https://github.com/salute-developers/golos
- Fanasr russo https://modelscope.cn/models/damo/speech_uniasr_asr_2pass-6k-common-vocab1664-tensorflow1-ffline/summary
Não testado (pior que a qualidade)
- https://github.com/sovse/base_rus_whisper_stt
Linguística (listas de palavras, morfologia)
- http://aot.ru
- https://natasha.github.io
Pontuação e cartas de título
https://alphacephei.com/vosk/models/vosk-recasepunch-en-0.22.zip
https://hugingface.co/kontur-a/sbert_punch_case_ru
https://github.com/kotikkontantin/ru-upopunction
https://github.com/vlomme/bert-russian-punctual
https://github.com/lesha17/punctual
https://github.com/gleb-skobinsky/ru_punct
https://github.com/sviperm/neuro-commma
https://github.com/snakers4/silero-models
https://github.com/marlon-rt/neuro-commma
https://github.com/sviperm/neuro-commma
https://github.com/averkij/multipunct
https://github.com/denis-berezutskiy-lad/transcription-bert-punctuator-scripts hugingface
https://hugingface.co/ai-forever/sage-fredt5-distille-95m-set de modelos SAGE
História
- A.A. Dicionário de gramática
- Otipple é um departamento e uma separação da linguística teórica e aplicada da Faculdade de Filologia da Universidade Estadual de Moscou
- MÉTODO LOBANOV BORIS
- IPPPI - Sorokin Victor Nikolaevich
- História do IPA Ras
- Processamento digital e reconhecimento dos sinais de fala do WC Ras
Alfacep
- 2005 começou a trabalhar no sintetizador do festival
- Festlang Clunits
- Idioma russo no voxforge
- Cmusphinx