https://github.com/earlephilhower/ESP8266SAM/issues/20#issue-1251795931

Originally posted by @wonsuchai in #20 (comment)

I have a PCM5102 connected to ESP8266 and all examples from ESP8266Audio work as expected. This means the wiring is right.

I can't get SAM's Say() to output anything. Any help appreciated.

I tried this on M5Stack Atom Echo Smart Speaker. The sound is very muffled. Is there anyway to tune the output?

Cool project!

It would be great if you had a video on YouTube showing this in action.

please delete

Could someone guide me on using the code on an esp32? I was using an Esp8266 with the nodac code. Using a single wire from rx to an input pin on my amp it works fine but once I switch to an Esp32 it doesn't seem to output anything on any Rx pins.
Oh and if anyone could teach me how to switch the voice presets in the code, that would be much appreciated as well. Best regards, H.

apparently pio is picking up the debug.x files and using them for all esp8266 projects, name conflict I guess with framework/debug.h

Hi.

I have a very specific use case. I hope to implement SAM, and put the output into the mic input of a walkie talkie. As such, I don't believe I need a DAC: I think the digital output and a capacitor will do the job. I'm not entirely sure which pin to use. The optical example in ESP8266Audio just says "GPIO pin" but I'm not sure ow to select this (if it can be selected).

The other thing is: I don't see how to use the SAM library. The instructions say I need https://github.com/earlephilhower/ESP8266Audio. Do I require to install ESP8266SAM as a library as well as ESP8266Audio?

Hope you can help, then (as requested) I believe I have a VERY interesting application for your wrapper!

Thanks.

Don't Ask Software morphed into SoftVoice, with their web site as http://text2speech.com/. You can see they note S.A.M. on the main page.

Not sure how that affects your readme, but thought the additional information might be of value.

Jim

When running the Code example ESP8266SAM->Examples->Speak theres noise in the background (more foreground):
Id attached a recorded example of it.

When using ESP8266 Audio and i play sth the noise does not appear.

noise.wav.zip

Do i have to configure sth to eliminate the noise in the foreground?

Similar to : earlephilhower/ESP8266Audio#473

On ESP32, after playing a SAM voice it never stop to say the last sound.
To find out this you have to modify a little the default example by replacing the delay(); with something not blocking.

Here the example to verify this little bug on ESP32 (and a quick/dirty solution in comments) :

#include <Arduino.h>
#include <ESP8266SAM.h>
#include <AudioOutputI2S.h>

unsigned long previousMillis = 0;        // will store last time sam voice was played
const long interval = 5000;              // inerval between the repeats

AudioOutputI2S *out = NULL;

void setup()
{
  out = new AudioOutputI2S();
  out->begin();
}

void loop()
{

  unsigned long currentMillis = millis();
  if (currentMillis - previousMillis >= interval) {
    previousMillis = currentMillis;
    ESP8266SAM *sam = new ESP8266SAM;
    sam->Say(out, "Bonjour");
    delete sam;
    // quick fix to avoird infinite repeat on ESP32 :
    //out->stop();  

  }
}

I hesitated to create this issue in ESP8266audio library but I think that the out->stop(); could be added at line 87 of ESP8266SAM.cpp.
Need expert opinion to confirm 😉

As per community topic

The debug.h header defines

With ESP8266 core also contains the debug.h header

https://github.com/esp8266/Arduino/blob/master/cores/esp8266/debug.h

In PlatformIO this causes a conflict due to probably a different order of include flags. An example of this is the WiFiManager library that has these functions

    void          setDebugOutput(boolean debug);
    void          setDebugOutput(boolean debug, String prefix); // log line prefix, default "*wm:"

with a previous #include <Arduino.h>, which at some points includes debug.h which then goes into this library's debug.h instead of the core's debug.h library, the #define debug 0 takes effect and the expanded version of the function declaration is

void          setDebugOutput(boolean 0);

which is invalid code.

In file included from C:\Users\Miguel\.platformio\packages\framework-arduinoespressif8266\cores\esp8266/Stream.h:25,
                 from C:\Users\Miguel\.platformio\packages\framework-arduinoespressif8266\cores\esp8266/HardwareSerial.h:32,
                 from C:\Users\Miguel\.platformio\packages\framework-arduinoespressif8266\cores\esp8266/Arduino.h:288,
                 from C:\Users\Miguel\AppData\Local\Temp\tmpvxy3fvbw:1:
.pio\libdeps\d1_mini\ESP8266SAM\src/debug.h:17:15: error: expected ',' or '...' before numeric constant
   17 | #define debug 0
      |               ^
.pio\libdeps\d1_mini\WiFiManager/WiFiManager.h:99:42: note: in expansion of macro 'debug'
   99 |     void          setDebugOutput(boolean debug);

I'd suggest to rename the debug.h file to esp8266sam_debug.h to avoid a conflict with Arduino-ESP8266 and also rename the debug to something more unique.

How do you use the different voices? Elf, Robot, etc.

When I add https://github.com/earlephilhower/ESP8266SAM/blob/master/library.json, I am unable to see the earlephilhower version of SAM. I can only see Juan Pablo Marquez' Spanish translation.

When I add https://github.com/earlephilhower/ESP8266Audio/blob/master/library.json, I am able to install ESP8266Audi, so I'm pretty sure I'm doing something right.

Any suggestions, please?

Hi, first of all, great work porting this library and on your other library ESP8266Audio as well 😃

I'm from Argentina, and I'm thinking it would be great to add Spanish support to ESP8266SAM. I think the existing phonemes should work, and the only thing to change would be the mathching from strings to phonemes, so it would implement Spanish pronounciation by selecting different phonemes than the ones for English. From playing with other libraries and frameworks I know this would not render optimal results, probably will sound a little funny, but maybe it could be a good starting point.

Would you provide any guidelines or ideas on how to achieve this? Thanks!

If you don't put a . after the speech text it makes a horrible screeching noise

Hi Phil
The SamD21 used in the Arduino Zero has an onboard dac (A0)
Any chance to have SAM running also on it?
Many thanks
Cheers,
Jean-Luc

Hi,
I think it would be good if you'd share a couple of audio examples of SAM saying a couple of sentences. Personally, it would help me in comparing my results with yours, because I'm not sure if the quality I'm getting is the best it could be.

thanks!

Hi there, and kudos for the excellent porting (and esp8266audio creation too).

I'm using the NoDAC + SAM successfully but now that the program is getting bigger, I noticed that code seems to not be able to loop anymore.
The only solution I've found so far, is to call SAM once per loop and just make him say nothing.
If I do this the other functions start working magically, like the server.handle() (for the incoming http connections to the webserver) and some simple timers;.

Is there something I am missing?
Any suggestion would be really appreciated.
Thank you ! :)

I am using the esp8266 (12) with NoDAC and get a foreground noise that renders the output (almost) unusable ... If I didn't know what it was saying ... I wouldn't ... I noticed that the piano sample in the esp8266audio library sounded the same way until I normalized, and amplified the mp3 ... any chance of making tweaks to the esp8266sam's output? I plan to test it with an i2c hardware decoder and amp to see if it's just the quality or the NoDAC setup ...

Hello Earle,

When trying out the Phonetic possibilities I noticed that SetPhonetic is not really implemented although mentioned.

I added the following lines to the file sam.c and then it appeared to work for real :

Line 17:
int phonetic = 0;

Line 61:
void SetPhonetic(int x) {phonetic = x;};

Phonetic Words as mentioned in the original Manual of 'Don't Ask Computer Software' do work now:
https://janderogee.com/projects/SerialSpeechSynthesisSAM/files/dict_eng.txt
e.g.
Q
quality=KWAA4LIXTIY
quantity=KWAA4NTIXTIY
quart=KWOH4RT
quarts=KWOH4RTS
quebec=KUHBEH4K
question=KWEH4SCHUN
quick=KWIH4K
quiet=KWAY4IXT
quit=KWIH4T
quiz=KWIH4Z
quote=KWOW4T
quotient=KWOW4SHUNT

void SAM_Phonetic_Test() {
char my_tts_text[150];
ESP8266SAM *sam = new ESP8266SAM;
sam->SetPhonetic(true); \\ Puts SAM in the Phonetic Mode

strncpy_P(my_tts_text, PSTR("KWAA4LIXTIY KWAA4NTIXTIY KWOH4RT KWOH4RTS KUHBEH4K KWEH4SCHUN KWIH4K KWAY4IXT KWIH4T KWIH4Z KWOW4T KWOW4SHUNT", sizeof(my_tts_text)));

sam->Say(audio_output_I2S, my_tts_text);
sam->SetPhonetic(false); \\ Puts SAM in the Normal Mode

sam->Say(audio_output_I2S, "Can you hear me now?");
delay(500);
sam->Say(audio_output_I2S, "I can't hear you!");

delete sam;

}

PS: Thanks for the work with great library, brings back sweet memories of my early experiences on my Apple-][ Clone

Hi. i could not test for now. still waiting for the correct hardware...

But can you tell me if it's possible to use esp8266sam without locking the main program loop ?

Does the main loop hang when the say method is called ?

Goal is to animate something on a screen while tts is running. could this be achieved ?

Thanks by advance for you answer.
DrZoid