Hi - I am working on converting Text to Speech using SSML via Google Speech API. Below is the request to the API. This perfectly works when language Code is En-US , however for code de-DE and to hear in German voice, the output is totally random. Please help me in checking this issue,
TTS Request JSON :: {"voice":{"ssmlGender":"MALE","name":"de-DE-Wavenet-E","languageCode":"de-DE"},"input":{"ssml":"<speak><say-as interpret-as=\"date\" format=\"yyyymmdd\"> 20220506<\/say-as><\/speak>"},"audioConfig":{"sampleRateHertz":8000,"volumeGainDb":0,"speakingRate":1,"audioEncoding":"LINEAR16","pitch":0,"effectsProfileId":["telephony-class-application"]}}
The same request when changeing the name and Language code works perfectly.
You will want to see this documentation where it’s explained how to use different languages instead of english in SSML they also provide some code examples as the one i’m sharing where it explains how to use the Voice tag in SSML.
<speak>The dog is friendly<voice name="fr-CA-Wavenet-B">mais la chat est mignon</voice><break time="250ms"/> said a pet shop owner</speak>
Hi Eduardo,
you may have a look at this as well. The google asr has still problems with german date transcription:
https://stackoverflow.com/questions/66799469/google-speech-recognition-weak-date-transcription
Best Regards
Andre
User | Count |
---|---|
2 | |
1 | |
1 | |
1 | |
1 |