Error with Japanese Language

[jeffs2303]

The latest version.

[ERRO] An unexpected error occurred while executing the task:
[ERRO] Unexpected error while executing task : Language 'jpn' is not supported by the selected TTS engine

I also -r="tts=macos|allow_unlisted_languages=True" but get [ERRO] Unexpected error while executing task : Both the C extension and the pure Python code failed. (Wrong arguments? Input too big?)

[And-0]

I'm on Windows, and I get exactly the same issues with espeak-ng. If I don't specify it, jpn (or jp or ja) is not supported; if I do specify -r="tts=espeak-ng|allow_unlisted_languages=True" I get "Both the C extension and the pure Python code failed. (Wrong arguments? Input too big?)"

I use the command:

python -m aeneas.tools.execute_task audio.m4a script.txt "task_language=jp|os_task_file_format=json|is_text_type=plain" -r="tts=espeak-ng|allow_unlisted_languages=True" map.json

Note that replacing jp with jpn or ja does not help, unfortunately.

[dndeus]

Any news on that?? I think it's a pretty major issue :)

[zhaozhuohang]

Any news on that?? I think it's a pretty major issue :)

[pettarin]

@And-0 : did you check that your underlying installation of espeak-ng can actually speak Japanese? See this comment #222 (comment) pointing to espeak-ng/espeak-ng#864 .

@jeffs2303 : did you check that the TTS provided by the macOS can actually speak Japanese? You might need to install additional languages in the macOS settings (apologies, but I no longer have a machine running macOS, so I cannot give precise instructions). Try running with the --verbose option (as done in #222 (comment) ) and see if the cause of the error becomes clearer.

[readbeyond]

@And-0 from the current HEAD of devel branch:

$ espeak --version
eSpeak text-to-speech: 1.48.15  16.Apr.15  Data at: /usr/lib/x86_64-linux-gnu/espeak-data

$ espeak-ng --version
eSpeak NG text-to-speech: 1.52.0  Data at: /usr/lib/x86_64-linux-gnu/espeak-ng-data

$ # m is an alias for micromamba --- feel free to use your preferred venv manager (conda, uv, etc.)
$ m env create -n test_aeneas_316 python=3.13
$ m activate test_aeneas_316

(test_aeneas_316) $ pip install setuptools wheel numpy

(test_aeneas_316) $ git status
On branch devel
Your branch is up to date with 'origin/devel'.

nothing to commit, working tree clean

(test_aeneas_316) $ git log -n1
commit 02d43937a965a3d9e067da9b9eccc6c32f119e69 (HEAD -> devel, origin/devel)
Merge: 48c1f8a bec9d7d
Author: ReadBeyond <5318774+readbeyond@users.noreply.github.com>
Date:   Sat Jul 25 12:22:31 2026 +0200

    Merge pull request #323 from pettarin/devel
    
    Fix warnings from aeneas_check_setup related to the escaping of backslash in u-strings

(test_aeneas_316) $ python setup.py build_ext --inplace
(test_aeneas_316) $ pip install -e .

(test_aeneas_316) $ pip list
Package           Version Editable project location
----------------- ------- ------------------------------------
aeneas            1.7.4.0 /home/alberto/projects/aeneas/aeneas
beautifulsoup4    4.15.0
lxml              6.1.1
numpy             2.5.1
packaging         26.2
pip               26.1.2
setuptools        83.0.0
soupsieve         2.9.1
typing_extensions 4.16.0
wheel             0.47.0

(test_aeneas_316) $ m list
List of packages in environment: "/home/alberto/micromamba/envs/test_aeneas_316"

  Name               Version    Build                 Channel    
───────────────────────────────────────────────────────────────────
  _openmp_mutex      4.5        20_gnu                conda-forge
  aeneas             1.7.4.0    pypi_0                pypi       
  beautifulsoup4     4.15.0     pypi_0                pypi       
  bzip2              1.0.8      hda65f42_9            conda-forge
  ca-certificates    2026.7.22  hbd8a1cb_0            conda-forge
  ld_impl_linux-64   2.46.1     default_hbd61a6d_102  conda-forge
  libexpat           2.8.1      hecca717_1            conda-forge
  libffi             3.5.2      h3435931_0            conda-forge
  libgcc             15.2.0     he0feb66_20           conda-forge
  libgomp            15.2.0     he0feb66_20           conda-forge
  liblzma            5.8.3      hb03c661_0            conda-forge
  libmpdec           4.0.0      hb03c661_1            conda-forge
  libsqlite          3.53.3     h0c1763c_0            conda-forge
  libuuid            2.42.2     h5347b49_0            conda-forge
  libzlib            1.3.2      h25fd6f3_2            conda-forge
  lxml               6.1.1      pypi_0                pypi       
  ncurses            6.6        hdb14827_0            conda-forge
  numpy              2.5.1      pypi_0                pypi       
  openssl            3.6.3      h35e630c_0            conda-forge
  packaging          26.2       pypi_0                pypi       
  pip                26.1.2     pyh145f28c_0          conda-forge
  python             3.13.14    h6add32d_100_cp313    conda-forge
  python_abi         3.13       8_cp313               conda-forge
  readline           8.3        h853b02a_0            conda-forge
  setuptools         83.0.0     pypi_0                pypi       
  soupsieve          2.9.1      pypi_0                pypi       
  tk                 8.6.13     noxft_hd70dff1_3      conda-forge
  typing_extensions  4.16.0     pypi_0                pypi       
  tzdata             2026c      h151e31d_0            conda-forge
  wheel              0.47.0     pypi_0                pypi       
  zstd               1.5.7      hb78ec9c_6            conda-forge

(test_aeneas_316) $ aeneas_check_setup 
[INFO] aeneas         OK
[INFO] ffprobe        OK
[INFO] ffmpeg         OK
[INFO] espeak         OK
[INFO] aeneas.tools   OK
[INFO] shell encoding OK
[INFO] aeneas.cdtw    AVAILABLE
[INFO] aeneas.cmfcc   AVAILABLE
[INFO] aeneas.cew     AVAILABLE
[INFO] All required dependencies are met and all available Python C extensions are working
[INFO] Enjoy running aeneas!

Now let's run a built-in example:

(test_aeneas_316) $ aeneas_execute_task --example-smil
[INFO] Running example task with arguments:
  Audio file:    aeneas/tools/res/audio.mp3
  Text file:     aeneas/tools/res/page.xhtml
  Config string: task_language=eng|is_text_type=unparsed|is_text_unparsed_id_regex=f[0-9]+|is_text_unparsed_id_sort=numeric|os_task_file_format=smil|os_task_file_smil_audio_ref=p001.mp3|os_task_file_smil_page_ref=p001.xhtml
  Sync map file: output/sonnet.smil
[INFO] Creating task...
( ... omitted a XMLParsedAsHTMLWarning not related to this issue ... )
[INFO] Creating task... done
[INFO] Executing task...
[INFO] Executing task... done
[INFO] Creating output sync map file...
[INFO] Creating output sync map file... done
[INFO] Created file 'output/sonnet.smil'

Let's try using -r="tts=espeak-ng" and task_language=jpn :

(test_aeneas_316) $ aeneas_execute_task aeneas/tools/res/audio.mp3 aeneas/tools/res/page.xhtml "task_language=jpn|is_text_type=unparsed|is_text_unparsed_id_regex=f[0-9]+|is_text_unparsed_id_sort=numeric|os_task_file_format=smil|os_task_file_smil_audio_ref=p001.mp3|os_task_file_smil_page_ref=p001.xhtml" output/sonnet.smil -r="tts=espeak-ng"
[INFO] Validating config string (specify --skip-validator to bypass)...
[INFO] Validating config string... done
[INFO] Creating task...
( ... omitted a XMLParsedAsHTMLWarning not related to this issue ... )
[INFO] Creating task... done
[INFO] Executing task...
[INFO] Executing task... done
[INFO] Creating output sync map file...
[INFO] Creating output sync map file... done
[INFO] Created file 'output/sonnet.smil'

The only catch at the moment is that if you use tts=espeak-ng, the C extension CEW wraps eSpeak, not eSpeak-ng, and therefore the eSpeak-ng TTS is called via subprocess, which is a bit slower than using the C extension.

Run with --verbose to see it:

(test_aeneas_316) $ aeneas_execute_task aeneas/tools/res/audio.mp3 aeneas/tools/res/page.xhtml "task_language=jpn|is_text_type=unparsed|is_text_unparsed_id_regex=f[0-9]+|is_text_unparsed_id_sort=numeric|os_task_file_format=smil|os_task_file_smil_audio_ref=p001.mp3|os_task_file_smil_page_ref=p001.xhtml" output/sonnet.smil -r="tts=espeak-ng" --verbose
...
[DEBU] Synthesizer: Selecting TTS engine...
[DEBU] Synthesizer: TTS engine: eSpeak-ng
[DEBU] ESPEAKNGTTSWrapper: No tts_path specified in rconf, setting default TTS path
[DEBU] ESPEAKNGTTSWrapper: TTS path is             espeak-ng
[DEBU] ESPEAKNGTTSWrapper: TTS cache?              False
[DEBU] ESPEAKNGTTSWrapper: Has Python      call?   False
[DEBU] ESPEAKNGTTSWrapper: Has C extension call?   False    <---
[DEBU] ESPEAKNGTTSWrapper: Has subprocess  call?   True     <---
[DEBU] ESPEAKNGTTSWrapper: Subprocess arguments: ['espeak-ng', '-v', 'VOICE_CODE_STRING', '-w', 'WAVE_PATH', 'TEXT_STDIN']
[DEBU] Synthesizer: Selecting TTS engine... done
...
( and for each text fragment: )
[DEBU] ExecuteTask: STEP 3 BEGIN (synthesize text)
[DEBU] Synthesizer: Synthesizing text...
[DEBU] ESPEAKNGTTSWrapper: Calling TTS engine via C extension or subprocess
[DEBU] ESPEAKNGTTSWrapper: C extension '' not recognized
[DEBU] ESPEAKNGTTSWrapper: Running the pure Python code
[DEBU] ESPEAKNGTTSWrapper: Synthesizing multiple via subprocess...
[DEBU] ESPEAKNGTTSWrapper: Calling TTS engine using multiple generic function...
[DEBU] ESPEAKNGTTSWrapper: Determining codec and sample rate...
[DEBU] ESPEAKNGTTSWrapper: Reading codec and sample rate from OUTPUT_AUDIO_FORMAT
[DEBU] ESPEAKNGTTSWrapper: Determining codec and sample rate... done
[DEBU] ESPEAKNGTTSWrapper:   codec:       pcm_s16le
[DEBU] ESPEAKNGTTSWrapper:   sample rate: 22050
[DEBU] ESPEAKNGTTSWrapper: Examining fragment 0 (no cache)...
[DEBU] ESPEAKNGTTSWrapper: Language to voice code: 'jpn' => 'ja'
[DEBU] ESPEAKNGTTSWrapper: Calling helper function
[DEBU] ESPEAKNGTTSWrapper: Synthesizer helper called with output_file_path=None => creating temporary output file
[DEBU] ESPEAKNGTTSWrapper: Temporary output file path is '/tmp/tmpzs_c3fv9.wav'
[DEBU] ESPEAKNGTTSWrapper: TTS engine reads text from stdin
[DEBU] ESPEAKNGTTSWrapper: Creating arguments list...
[DEBU] ESPEAKNGTTSWrapper: Creating arguments list... done
[DEBU] ESPEAKNGTTSWrapper: Calling TTS engine...
[DEBU] ESPEAKNGTTSWrapper: Calling with arguments '['espeak-ng', '-v', 'ja', '-w', '/tmp/tmpzs_c3fv9.wav']'    <---
[DEBU] ESPEAKNGTTSWrapper: Calling with text 'I'
[DEBU] ESPEAKNGTTSWrapper: Passing text via stdin...
[DEBU] ESPEAKNGTTSWrapper: Passing text via stdin... done
[DEBU] ESPEAKNGTTSWrapper: TTS engine wrote audio data to file
[DEBU] ESPEAKNGTTSWrapper: Calling TTS ... done
...

I will address this performance issue in #324 .