Version

feat-dlspeech or master

Describe the bug

Safari requires audio buffer of 22100 Hz or larger. But Cognitive Services return PCM at 16000 Hz only.

We need to upscale it to 22100 Hz or 32000 Hz.

Steps to reproduce

1. On iOS, try out Direct Line Speech sample
2. Say "What's the weather" to the bot

Expected behavior

The bot should synthesize this "This is the forecast for this week".

Additional context

[Bug]