Speaking issues Coqui TTS Tacotron2 DDC model: Difference between revisions
Jump to navigation
Jump to search
(Restructure and add section for mispronounced words) |
|||
| Line 9: | Line 9: | ||
Most models are trained with a dot, exclamation or question mark at the end. So always end a sentence to avoid model synthesizing weird output. | Most models are trained with a dot, exclamation or question mark at the end. So always end a sentence to avoid model synthesizing weird output. | ||
== | ==Input string formatting== | ||
"ah" at end of sentence generally produces strange results. | ===Phrases ending in "ah"=== | ||
"ah" at end of sentence generally produces strange results. Short names produce a 12 second clip. | |||
Examples: | |||
== | * Nelson Mandela | ||
To speak acronyms as letters it needs to be formatted as "A. B. C. news" | * pergola | ||
==== Mitigation ==== | |||
If at the end of the input, adding punctuation to the end synthesizes correctly: | |||
Example "Nelson Mandela" > "Nelson Mandela." | |||
=== Acronyms === | |||
To speak acronyms as letters it needs to be formatted as:<blockquote>"A. B. C. news"</blockquote>Not:<blockquote>"ABC news" | |||
"A.B.C. news"</blockquote> | |||
== Mispronounced Words == | |||
* video | |||
Revision as of 02:25, 28 October 2021
- TTS model: tts_models/en/ljspeech/tacotron2-DDC
- Vocoder model: vocoder_models/en/ljspeech/hifigan_v2
General
Most models are trained with a dot, exclamation or question mark at the end. So always end a sentence to avoid model synthesizing weird output.
Input string formatting
Phrases ending in "ah"
"ah" at end of sentence generally produces strange results. Short names produce a 12 second clip.
Examples:
- Nelson Mandela
- pergola
Mitigation
If at the end of the input, adding punctuation to the end synthesizes correctly:
Example "Nelson Mandela" > "Nelson Mandela."
Acronyms
To speak acronyms as letters it needs to be formatted as:
"A. B. C. news"
Not:
"ABC news" "A.B.C. news"
Mispronounced Words
- video