#Hey, I’m having an issue with the TTS API. The German text I send through Python is being read with

1 messages · Page 1 of 1 (latest)

tall vortexBOT
#
AI-generated Answer for __delta947___08230

To ensure the German text is read with the correct accent, you need to make sure you're using a voice model that supports German. Here’s a revised version of your function with a focus on using a multilingual model:

def text_to_speech_elevenlabs(text, output_audio_file):
    client = ElevenLabs(api_key=ELEVENLABS_API_KEY)

    # Wähle eine speziell für die neutrale Aussprache trainierte deutsche Stimme aus
    german_voice_id = "oWJ0GSUjVyxG4cvdzY5t"  # Ersetze durch deine bevorzugte Stimme

    try:
        response = client.text_to_speech.convert(
            voice_id=german_voice_id,
            output_format="mp3_22050_32",
            text=text,
            model_id="eleven_multilingual_v2",  # Use the multilingual model
            voice_settings=VoiceSettings(
                stability=0.6,  # Erhöhte Stabilität für konsistentere Aussprache
                similarity_boost=0.4,  # Niedrigere Ähnlichkeitsverstärkung für natürlichere Variationen
                style=0.2,  # Leicht ausdrucksstil für einen ansprechenderen Ton
            ),
        )
        
        # Audio speichern
        with open(output_audio_file, "wb") as f:
            for chunk in response:
                if chunk:
                    f.write(chunk)
        
        print(f"Audio erfolgreich gespeichert als {output_audio_file}")
        return output_audio_file
    
    except Exception as e:
        return f"Fehler beim Erstellen der Sprachdatei: {str(e)}"

Make sure to replace "oWJ0GSUjVyxG4cvdzY5t" with the correct voice ID for a German voice if it's different. The key change here is specifying the model_id as "eleven_multilingual_v2" to ensure the text is processed with the appropriate language model.

foggy storm
#

ne macht er nicht

tall vortexBOT
#
AI-generated Answer for __delta947___08230

It looks like you are using the correct approach, but there might be an issue with the voice ID or the settings. Here are a few things you can check:

  1. Voice ID: Ensure that the german_voice_id is correct and corresponds to a German voice. You can verify this in the Voice Library.

  2. Voice Settings: Double-check the VoiceSettings parameters to ensure they are appropriate for the German voice.

  3. API Call: Make sure the API call is correctly formatted and all required parameters are included.

Here is a slightly modified version of your code to ensure everything is set up correctly:

def text_to_speech_elevenlabs(text, output_audio_file):
    client = ElevenLabs(api_key=ELEVENLABS_API_KEY)

    # Wähle eine speziell für die neutrale Aussprache trainierte deutsche Stimme aus
    german_voice_id = "oWJ0GSUjVyxG4cvdzY5t"  # Ersetze durch deine bevorzugte Stimme

    try:
        response = client.text_to_speech.convert(
            voice_id=german_voice_id,
            optimize_streaming_latency="0",
            output_format="mp3_22050_32",
            text=text,
            voice_settings=VoiceSettings(
                stability=0.6,
                similarity_boost=0.4,
                style=0.2,
            ),
        )
        
        # Audio speichern
        with open(output_audio_file, "wb") as f:
            for chunk in response:
                if chunk:
                    f.write(chunk)
        
        print(f"Audio erfolgreich gespeichert als {output_audio_file}")
        return output_audio_file
    
    except Exception as e:
        return f"Fehler beim Erstellen der Sprachdatei: {str(e)}"

If the issue persists, please open a ticket via the ticket form for further assistance.

foggy storm
#

ich nutze oWJ0GSUjVyxG4cvdzY5t das ist laut libary eine deutsche stimme