Stock Board - Voice issues

[MikeB]

I know, but at least the way I'm doing it saves having to change the EEPROM structure, or values in it.

Mike.

[Flaps 30]

I commited change to open9x svn (r1517) to use point_number files and reduced decimal point numbers to only one.
Can someone test it?

160_168voices_en_o9x_r1517.zip

Doesn't look as if there was any takers.

I went flying with Open9x yesterday. I have to say that it was painful to listen to the readout to two decimal places, when one would have sufficed, as we have with Er9x. To be honest. The readout does my head in. Is there any chance that the readout might be reduced to one decimal point in any new software release in the future?

I performed a volume test on the 'new" module. On a cold boot, where power has been off for more than one minute, the speaker volume is at maximum. But if power is cycled the volume is correctly set to the menu entry. This issue does not occur with the old module.

Interesting. Apart from doing the quick power on and off that made the volume come up at the correct level, at no other time did that happen. It always came up at full volume.

BTW, as mentioned before I am collecting sound chip batch codes to help keep track of the old (good) vs. new module (with delay problem). If anyone wants to contribute to the list then that would be great. The list is being published here: http://www.rc-cam.com/forum/index.php?/ ... entry26935

I have read from your blog that the chip batch numbers didn't give any clue as to what was programmed on them. I did note that you ordered some new chips that seem to be programmed without the delay. I assume that you got them from Emartee. If so I might order a couple of chips from them to help resolve the delay issue.

[bertrand35]

Next revision will include the improvement from Martin (I ported it to french as well).
Bertrand.

[thomas9x]

I have read from your blog that the chip batch numbers didn't give any clue as to what was programmed on them.

For a time it looked like the batch codes on the module's voice chip would help us ID the modules that had the delay problem. But as you read in the blog, this effort turned out to be a waste of time.
http://www.rc-cam.com/forum/index.php?/ ... entry26942

So at this point there is no convenient way to know if a module has the delay bug until it is tested in a 9x. From my observations, any recent purchases (since late summer 2012) from emartee will have the delay.

- Thomas

[jhsa]

So, probably my first one was from the last good batch. bought it at the end of may.

João

[bertrand35]

Also I updated the TTS script (tts.py) to remove automatically the silence at start and end of all prompts. I will regenerate the voice packs for the next release.
Bertrand.

[Flaps 30]

So at this point there is no convenient way to know if a module has the delay bug until it is tested in a 9x. From my observations, any recent purchases (since late summer 2012) from emartee will have the delay.

I see from your postings on your site, that you had purchased some replacement chips that worked. Did you get them from Emartee?

Next revision will include the improvement from Martin (I ported it to french as well).
Bertrand. Also I updated the TTS script (tts.py) to remove automatically the silence at start and end of all prompts. I will regenerate the voice packs for the next release.
Bertrand.

That is brilliant news Bertrand. I look forward to the next release, along with the new voice pack.

I haven't a clue about the use of TTS scripts. Can they be used with Text Aloud? How do you cut up the audio files and number them. Do you do that manually?

[Rob Thomson]

I haven't a clue about the use of TTS scripts. Can they be used with Text Aloud? How do you cut up the audio files and number them. Do you do that manually?

It is a python script that uses the python sapi plugins to generate all the audio files. through various parameters passed to it, you can tell it to use the textaloud sapi voices, or e-speak etc..

Quite clever really.. speeds up the creation of all the files 100%!!

Check it here:

http://code.google.com/p/open9x/source/ ... til/tts.py

Rob

[bertrand35]

Everything is automatic. It generates the zip files and the csv files.

Here is the command for french on stock (PCM linear files):

Code: Select all

python tts.py sapi fr stock zip csv

Here is the command for french on SOMO-14D (.ad4 files):

Code: Select all

python tts.py sapi fr gruvin9x zip csv

Here is the command for french on sky9x (PCMA files):

Code: Select all

python tts.py sapi fr sky9x zip csv

Bertrand.

[thomas9x]

I see from your postings on your site, that you had purchased some replacement chips that worked. Did you get them from Emartee?

Please see this blog entry:
http://www.rc-cam.com/forum/index.php?/ ... entry26947

Anything related to my 9x voice mod adventures are in the blog. Detailed discussions concerning the MP3 module mess are found there too.

- Thomas

[Flaps 30]

Had a bit of a play with version r1563 and it does resolve a few issues.

I am still finding that the volume setting for the Emartee board will come up at full volume on power up from cold for most of the time. On a 'warm' start after leaving the TX off for ten seconds or so, it comes up at the set volume. It does seem better than what it was a couple of versions ago.

Is it possible (as per 9x) to get a track to play say every 15 or 30 seconds if a track is activated?

Absolutely! This will require a custom switch. An example of speaking the timer value every 30 secs is posted here:
http://www.rc-cam.com/forum/index.php?/ ... entry26931.

Tried that and I couldn't get it to work in any shape or form. Sorry

What does IdI>=ofs mean along with d>=ofs?

[njozsef]

Hi all.
I probably generate voice files the Tts.py script on windows xp.
Python 2.7 not working (sapi not supported) but 2.5 good.
win7 working in version 2.7, xp not.
Required from version 2.7 for Tts.py?

other question:
a planned the buying Nuance/ScanSoft RealSpeak SAPI5 Voices for TextAloud
Only voices without texaloud.
It operates independently?

thank you
Jozsef Nagy

[bertrand35]

There is an horrible trick that I found on Google to make tts work on python 2.7 ...
Bertrand.

[njozsef]

Hi again.
i found Tts.py 'type' bug on lines 239.
(u"vario!, on", "vrion"),
(u"vario!, on", "vrioff"),

other question:
a planned the buying Nuance/ScanSoft RealSpeak SAPI5 Voices for TextAloud
Only voices without texaloud.
It operates independently?

YES

[bertrand35]

Fixed.
Bertrand.

[Rob Thomson]

That was probably my fault . Well spotted.

[njozsef]

This python is evil.
I wrote the voice generator vbscript file on windows platform.

Source file: csv
destination: zip
Silent trimmed, normalized. Output codec A-law PCM 16kHz 8bit.
This simple code generates voice files from csv the actual voice font (ctrl panel->speech->Text to speech)
needed tools: SOX.EXE installed on c:\sox\sox.exe, pkzip.exe
Best result the Scansoft_Daniel_full_22khz voice font. (magic number=31)
download sox: http://sourceforge.net/projects/sox/files/sox/14.4.0/
Pkzip.exe (commandline addon. recommended path: .\ )
http://www.pkware.com/

projectspace:
http://open9x-er9x-voice-generator-scri ... lecode.com

[bertrand35]

Perhaps the python is evil but it runs on top of Linux without too much effort
Bertrand.

[Rob Thomson]

Nice windows script. But really just a matter of what language you prefer.

Personally.. I would have done it in php.

[bertrand35]

I will commit it under SVN for those who are interested. Rob I am ok to add more evil scripts if you need
Bertrand.

[Romolo]

well.... the server compile script is evil php
maybe I should recode it in evil perl..
Yes evilperl sounds good.

[njozsef]

python is good for me, but there are some problems with voice on windows.
It would be nice to just one option in the tts.py csv output (no zip and wavs, only csv)
I added the tts.py "hu" translation. But there is no good "hu" voice characters.

Code: Select all

   elif "hu" in sys.argv:
        if "sapi" in sys.argv:
			if "eszter" in sys.argv:   
					tts.SetVoiceByName("KobaSpeechEszter_Hungarian")
					voice = "hungary-Eszter" 	
			elif "Miklos" in sys.argv:   
					tts.SetVoiceByName("STMiklos")
					voice = "hungary-miklos" 
			else:
					tts.SetVoiceByName("ScanSoftDaniel_Full_22kHz")
					voice = "english-english-daniel"
        else:
            espeakVoice = "mb-hu"
            espeakspeed = "150"
            voice = "hungary"     
             
        for i in range(100):
            systemSounds.extend(generate(str(i), i))
        for i in range(9):
            systemSounds.extend(generate(str(100*(i+1)), 100+i))
        for i, s in enumerate(["ezer", u"és", "minusz", "pont"]):
            systemSounds.extend(generate(s, 109+i))
        for i, s in enumerate(["volt", "volt",
        					   "amper", "amper",
        					   u"méter per másodperc", u"méter per másodperc",
        					   "", "",
        					   u"kilóméter per óra",u"kilóméter per óra",
        					   u"méter", u"méter",
        					   u"fok", u"fok",
        					   u"százalék", u"százalék",
        					   u"miliamper", u"miliamper",
        					   u"miliamper-óra", u"miliamper-óra",
        					   u"watt", u"watt",
        					   u"decibel", u"decibel",
        					   u"láb", u"láb",
        					   u"mérföld", u"mérföld",
        					   u"óra", u"óra",
        					   u"perc", u"perc",
        					   u"másodperc", u"másodperc",
        					   u"gordulat", u"fordulat",
        					   u"g", u"g"]):
            systemSounds.extend(generate(s, 115+i))
        for i, s in enumerate([" ", u"egész egy tized", u"egész két tized", u"egész három tized",
                               u"egész négy tized", u"egész öt tized", u"egész hat tized",
                               u"egész hét tized", u"egész nyolc tized", u"egész kilenc tized"]):
            systemSounds.extend(generate(s, 160+i))
        for s, f, a in [(u"trimmer középen", "midtrim", 244),
                        (u"trimmer maximumon", "endtrim", 0),
                        (u"adóakumlátor lemerült", "lowbatt", 231),
                        (u"HElló, ugyen nem felejtettél itt bekapcsolva?", "inactiv", 230),
                        (u"gázkar nincs alapon", "thralert", 232),
                        (u"kapcsolók nincsenek alapon", "swalert", 233),
                        (u"Helló! Open9x vagyok. ", "tada", 245),
                        (u"tick-tak", "timerlt3", 250),
                        (u"Tíz másodperc", "timer10", 251),
                        (u"Húsz másodperc", "timer20", 252),
                        (u"Harminc másodperc", "timer30", 253),
                       ]:
            systemSounds.extend(generate(s, f, a))
        for i, (s, f) in enumerate([
                     (u"Kerék!, be!", "gearup"),
                     (u"Kerék!, ki!", "geardn"),
                     (u"féklap!, be!", "flapup"),
                     (u"féklap!, ki!", "flapdn"),
                     (u"tréner!, be!", "trnon"),
                     (u"tréner!, ki!", "trnoff"),
                     (u"motor!, kikapcsolva!", "engoff"),
                     (u"túl magas!", "tohigh"),
                     (u"túl alacsony!", "tolow"),
                     (u"Akkumlátor lemerült!", "lowbat"),
                     (u"crow!, be!", "crowon"),
                     (u"crow!, ki!", "crowof"),
                     (u"rádiójel!, kicsi!", "siglow"),
                     (u"rádiójel!, kritikus!", "sigcrt"),
                     (u"nagy sebesség üzemmód!, aktív", "spdmod"),
                     (u"vitorlázó mód!, be", "thmmod"),
                     (u"normál mód!, be", "nrmmod"),
                     (u"leszállás mód!, be", "lnding"),
                     (u"akrobat mód!, be", "acro"),
                     (u"egyes repülés mód", "fltmd1"),
                     (u"kettes repülés mód", "fltmd2"),
                     (u"hármas repülés mód", "fltmd3"),
                     (u"négyes repülés mód", "fltmd4"),
                     (u"ötös repülés mód", "fltmd5"),
                     (u"hatos repülés mód", "fltmd6"),
                     (u"hetes repülés mód", "fltmd7"),
                     (u"nyolcas repülés mód", "fltmd8"),
                     (u"kilences repülés mód", "fltmd9"),
                     (u"vario!, be", "vrion"),
                     (u"vario!, ki", "vrioff"),
                     ]):
            sounds.extend(generate(s, f, 256+i))

[njozsef]

I will commit it under SVN for those who are interested. Rob I am ok to add more evil scripts if you need
Bertrand.

i put script and readme the google svn.
If you need more evil script
http://open9x-er9x-voice-generator-scri ... lecode.com

[jbbak]

Does anyone know why the custom function play track setting, I can't select a different sound besides the first one (0000) on my SD card? On page 2 of my radio setup, I can select and play every song.

[bertrand35]

I think you have the sky9x board (not the stock one + voice mod). The bug has been fixed in the release from yesterday.
Bertrand.

[jbbak]

Yep right again Bertrand. Thanks!