Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

[Bug] espeak-ng produces language flags

See original GitHub issue

Describe the bug

Espeak-ng phonemizer sometime produces language flags based on language rulesets. Those flags are kept by the current implementation and forwarded upstream to the tokenizer inducing phoneme mismatches.

To Reproduce

from TTS.tts.utils.text.phonemizers.espeak_wrapper import ESpeak
e = ESpeak(language="fr-fr")
e.phonemize("Certains mots comme football génèrent des flags de langue", separator="")
e = ESpeak(language="de")
e.phonemize("mein neues bike", separator="")

'sɛʁtˈɛ̃ mˈo kɔm (en)fˈʊtbɔːl(fr) ʒenˈɛʁ de- flˈaɡ də- lˈɑ̃ɡ'
#---------------^^^^--------^^^^
'maɪn nˈɔøəs (en)bˈaɪk(de)'
#------------^^^^-----^^^^

Expected behavior

from TTS.tts.utils.text.phonemizers.espeak_wrapper import ESpeak
e = ESpeak(language="fr-fr")
e.phonemize("Certains mots comme football génèrent des flags de langue", separator="")
e = ESpeak(language="de")
e.phonemize("mein neues bike", separator="")

'sɛʁtˈɛ̃ mˈo kɔm fˈʊtbɔːl ʒenˈɛʁ de- flˈaɡ də- lˈɑ̃ɡ'
'maɪn nˈɔøəs bˈaɪk'

Logs

No response

Environment

{
    "CUDA": {
        "GPU": [],
        "available": false,
        "version": "10.2"
    },
    "Packages": {
        "PyTorch_debug": false,
        "PyTorch_version": "1.12.0+cu102",
        "TTS": "0.7.1",
        "numpy": "1.21.6"
    },
    "System": {
        "OS": "Linux",
        "architecture": [
            "64bit",
            "ELF"
        ],
        "processor": "",
        "python": "3.9.13",
        "version": "#1 SMP Debian 5.10.127-1 (2022-06-30)"
    }
}

eSpeak NG text-to-speech: 1.50

Additional context

Pull request available at #1801

Issue Analytics

State:
Created a year ago
Comments:5 (4 by maintainers)

Top GitHub Comments

1reaction

erogolcommented, Aug 1, 2022

@WeberJulian There is already a ✨PR✨ #1801

0reactions

erogolcommented, Aug 15, 2022

Fixed by #1801

Top Results From Across the Web

Emoji support produces incomplete or corrupt translations #308

Based on the previous bug, we can understand that that "sign" is dropped because perhaps the cause of this bug happens so that...

eSpeak NG Text-to-Speech - Google Git

The eSpeak NG (Next Generation) Text-to-Speech program is an open source speech synthesizer that supports 102 languages and accents, based on the eSpeak ......

How to Add a Language to eSpeak NG

Installing eSpeak NG. First, let's clone eSpeak NG from Github: josh@yoga:~/Desktop$ git clone https://github.com ...

1376900 – Review Request: espeak-ng - Red Hat Bugzilla

Bug 1376900 - Review Request: espeak-ng - eSpeak NG Text-to-Speech ... is an open source speech synthesizer that supports over 70 languages.

Why flags do not represent languages

Flags are unique to a country or nation: but languages are often spoken across ... The Spanish flag is often used to represent...