My Azure Text-to-speech app no longer outputs once i added the ssml string

Question

I created a text-to-speech script using Microsoft Azure. Today I decided i wanted to add a pitch changer, speaking rate changer and possibly some silence additions. For that, i need to replace my speak_text_async(text) with speak_ssml_async(ssml_string). Ever since I did that, the TTS stopped playing and no .wav files are being generated. All i did was add the constant 50% pitch just to test it out, the ssml_string and changed the synthesizer to ssml instead of text (otherwise it would just read the html lines within the ssml.

I only changed the speak_ssml_async back to speak_text_async to but kept (ssml_string) to confirm that the problem is from the ssml_string but I can't figure out what it is because i receive no errors.

I will leave here the relevant part of my code. Keep in mind that i have custom output filename and directory selectors, as well as an input label for the tts text input before this def.

        #Directory selector
        output_label = ttk.Label(self, text="Choose your output folder:", 
                                 font=platformfont, 
                                 style="Output.TLabel")
        output_label.pack(pady=2)
        self.output_dir_button = ttk.Button(self, text="Browse", command=self.choose_output_dir, 
                                            takefocus=False, 
                                            style="Custom.TButton")
        self.output_dir_button.pack()
        self.output_dir_path = tk.StringVar()
        self.output_dir_path.set("")
        self.output_dir_entry = tk.Entry(self, textvariable=self.output_dir_path, font=inputfont, 
                                        width=55, 
                                        foreground="#395578", 
                                        state='readonly', 
                                        background="light gray", 
                                        readonlybackground="#Eed9c9", 
                                        borderwidth=0, 
                                        cursor="X_cursor",
                                        relief="flat")
        self.output_dir_entry.pack(pady=5)

        #Output filename
        output_filename_label = ttk.Label(self, text="Enter output filename (without extension):", 
                                          font=platformfont, 
                                          style="Output.TLabel")
        output_filename_label.pack(pady=5)

        #Listen button
        speak_button = ttk.Button(self, text="Listen & Generate", command=self.speak_text, 
                                  takefocus=False, 
                                  style="Custom.TButton")
        speak_button.pack(pady=15)

    def choose_output_dir(self):
        dir_path = filedialog.askdirectory()
        if dir_path:
            self.output_dir_path.set(dir_path)

    def speak_text(self):
        text = self.input_text.get("1.0", "end")
        output_dir = self.output_dir_path.get()
        output_filename = self.output_filename_text.get()
        if output_filename == "":
            output_filename = "tcnoutput"
        output_file = os.path.join(output_dir, output_filename + ".wav")

        if os.path.exists(output_file):
            response = messagebox.askyesnocancel("File Exists", "A file with the same name already exists. Do you want to overwrite it?",
                                            icon='warning')
            if response == True:
                os.remove(output_file)
            elif response == False:
                i = 1
                while os.path.exists(os.path.join(output_dir, output_filename + f"({i})" + ".wav")):
                    i += 1
                output_filename = output_filename + f"({i})"
                output_file = os.path.join(output_dir, output_filename + ".wav")
            else:
                raise KeyboardInterrupt
            
        pitch = "+50.0%"
        ssml_string = f"<speak version='1.0' xmlns='http://www.w3.org/2001/10/synthesis' xml:lang='ro-RO'>" \
                      f"<prosody pitch='{pitch}'>{text}</prosody></speak>"
        
        speech_synthesis_result = self.speech_synthesizer.speak_ssml_async(ssml_string).get()
        if speech_synthesis_result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
            with open(output_file, "wb") as f:
                f.write(speech_synthesis_result.audio_data)
            if output_dir == "":
                output_final = os.getcwd() + "\\" + output_filename + ".wav"
            else:
                output_final = output_dir + "/" + output_filename + ".wav"
            messagebox.showinfo("Success", f"Audio file successfully saved at: {output_final}")
        else:
            messagebox.showerror("Error", "Speech synthesis failed.")

Answer 1

I tried the below python code to configure the text to speech and configure speech settings with SSML and got the desired audio output like below:-

Code:

import  azure.cognitiveservices.speech  as  speechsdk
import  io
import  wave
speech_config = speechsdk.SpeechConfig(subscription="key", region="region")
synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config)
ssml_string = "<speak version='1.0' xmlns='http://www.w3.org/2001/10/synthesis' xml:lang='en-US'><voice name='en-US-JennyNeural'><prosody pitch='+50%'>Hello, my friend! How are you?</prosody></voice></speak>"
result = synthesizer.speak_ssml_async(ssml_string).get()
if  result.reason != speechsdk.ResultReason.SynthesizingAudioCompleted:
    print("SSML string is incorrect: {}".format(result.errorDetails))
else:
    print("SSML string is correct")
    with  io.BytesIO(result.audio_data) as  compressedAudioStream:
        with  wave.open("test.wav", "wb") as  wavFile:
            wavFile.setnchannels(1)
            wavFile.setsampwidth(2)
            wavFile.setframerate(16000)
            wavFile.writeframes(compressedAudioStream.read())

Output:
The audio of the input text is generated in wav file ,

• Refer these 2 MS documents to configure SSML code Doc1 &Doc2