Tjiho
/
French-tts-model-piper
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
French TTS models for pipers
6 Commits 1 Branch 0 Tags 172 MiB
Go to file
Tom Darboux 1b6757eb2a Improve readme 2023-11-03 20:12:03 +01:00
model 1 Add two TTS models 2023-10-26 23:21:01 +02:00
model 2 Add two TTS models 2023-10-26 23:21:01 +02:00
model 3 Add third model 2023-11-03 19:59:12 +01:00
LICENSE Change license to AGPL 2023-10-30 23:51:28 +01:00
README.md Improve readme 2023-11-03 20:12:03 +01:00
long-example-1.wav Add two TTS models 2023-10-26 23:21:01 +02:00
long-example-2.wav Add two TTS models 2023-10-26 23:21:01 +02:00
small-example-1.wav Add two TTS models 2023-10-26 23:21:01 +02:00
small-example-2.wav Add two TTS models 2023-10-26 23:21:01 +02:00
small-example-3.wav Add third model 2023-11-03 19:59:12 +01:00

README.md

French-tts-model-piper

French TTS models for piper

The first model is trained both on https://git.bksp.space/Tjiho/newspaper-sentences and https://git.bksp.space/Tjiho/baudelaire-sentences.

The second one is only trained on https://git.bksp.space/Tjiho/baudelaire-sentences.

The third is trained 300 epochs longer than the previous one with the same dataset.

You can use them with Piper speech synthetiser https://github.com/rhasspy/piper

echo "Je suis un modèle d'intelligence artificielle qui parle français" | ~/bin/piper/piper -m <model> --output_file <your_file>

Both are trained on 44100hz audio.