[hts-users:03682] Re: Voice frequency
- Subject: [hts-users:03682] Re: Voice frequency
- From: Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
- Date: Sat, 30 Mar 2013 19:45:05 +0900
- Cc: uratec <uratec@xxxxxxxxxxxx>
- Delivered-to: hts-users@xxxxxxxxxxxxxxx
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; bh=9BKZFHG7myqsm7ZHKdyE2O/92476OzgDaXBhTVzpK14=; b=ORShQRukvZKG4fz69wU053u4Mu1FUSevk1t4pR7FpNSKNf/PBMbPCB/IXVlVf+h0gL Bl+Y49CnAE8B90p0Ez0WUKujwEoXuZhR6YJ0P6uaF0JHXCSBLr5nnf4cNtQ+/9fll0jA yccdV0/IOJo+g6ki0jH78+IH3udtwr7EgY2iUO375/NrLbLWyzf2K1e44+LPeNwFFGSo E4Cfdo2qNAx86CxoUssWTFBDA/6CRVOEW2Yvt4z29+EvbZaZ1/6Oglk9ih7QUh9KuGID kDy0L6Vy7jo8B6Xi/vdIZQajwoFvHDW3mMIk9HdiORYtlrHNvl5FvxXGePJ3Wk3SRXCR Isjg==
Hi,
In HTS demo scripts, F0 is modeled in Log Hz.
Log F0 sequence of training data and generated data can be shown as follows.
x2x +fa data/lf0/cmu_us_arctic_slt_a0001.lf0
x2x +fa gen/qst001/ver1/1mix/0/alice01.lf0
The log F0 distribution of each full-context model is stored in
models/qst001/ver1/cmp/re_clustered.mmf.
Regards,
Keiichiro Oura
2013/3/27 David Goldsmith <david@xxxxxxxxxx>:
> Hello,
> I am using HTS in a tool which needs to be able to change the base tone
> of the speech by frequency, ie, "+10Hz", etc... As far as I understand,
> HTS only handles in half-tones, correct? Is it possible to get the
> fundamental frequency of the currently loaded voice model? If so, I
> could simply convert the input hz value into a corresponding semi-tone
> value for HTS. Or is there possibly a better way I am not aware of?
> Thank you for your assistance.
>
>
- References
-
- [hts-users:03680] Voice frequency, David Goldsmith